Skip to main content

Full text of "Muhammad At Medina"

See other formats



0SMAJ& ' 

Author VO^t^i IO 

Title rtuJka wivv^JL A^ l?e^UnCi ". 

*Thls1)o6k should be returned on or before the date last marked Wlow. 






Oxford University Press, Amen House, London E.G. 4 


Geoffrey Gumberlege, Publisher to the University 



THE present volume is a sequel to Muhammad at Mecca, and 
the two together are intended to constitute a history of 
the life of Muhammad and of the origins of the Islamic 
community. The plan of the book should be clear from the table 
of contents. I have endeavoured to write so as to be easily under- 
stood by the historian who has no knowledge of Arabic, but I have 
probably often fallen short of this aim. In particular, in discussions 
of a pioneering character, such as those in the fourth and fifth 
chapters, I have necessarily written at greater length than the 
intrinsic importance of the topic warranted, and thereby upset the 
balance of the various p^rts. In such cases all I can do is to advise 
the non-specialist to 'skip* judiciously. 

In a subject like that of this book where there, is a vast mass 
both of source material and of scholarly discussions, it is difficult 
not to overlook points here and there. I trust, however, that 
nothing of importance has been omitted. The exhaustive treatment 
of a subject is a noble ideal to have before one's eyes, but in 
scholarship as in economics the law of diminishing returns is 
operative. A point is reached at which further heavy labour leads 
to a negligible improvement in the product. While few readers 
are likely to be as fuUy aware as I am of the places where further 
study is possible, I have decided that, for the moment at least, I 
have said my say about Muhammad, and, if I try to say more, am 
as likely to mar as to better the impression I have tried to convey. 

It is appropriate at this point to draw attention to two gaps of 
which I have become aware in the course of my work, and which 
the normal type of European or American orientalist is incapable 
of filling. One is the production of a nlap of Arabia as it was in 
Muhammad's time. For this the information to be gathered from 
the old Arab geographers has to be transferred to a series of large- 
scale modern maps of the country; and that can hardly be done 
without access to all the localities. An excellent beginning has 
been made by one Muslim scholar, 1 and it is to be hoped that 
others will continue the work. 

The other serious gap is that the study of life in pre-Islamic 

1 Cf. p. i%3 below. 


Arabia has not kept pace with the development of social anthro- 
pology. I have done what I could to fill in this gap in so far as 
pre-Islamic conditions are necessary as a background for an under- 
standing of Muhammad's social reforms. From my colleague, 
Dr. Kenneth L. Little, head of the Department of Social Anthro- 
pology in the University of Edinburgh, I received valuable help, 
and I am much indebted to him for enabling me to correct some 
elementary mistakes. My fumbling attempts, however, have con- 
vinced me that the adequate study of pre-Islamic life demands 
someone who is primarily a social anthropologist, but who is at 
the same time able to deal directly with the Arabic source material. 
The non-anthropologist inevitably overlooks the significance of 
many details in the material. 

The transliteration of Arabic names is the same as in Muhammad 
at Mecca with one small exception. Where two 'letters indicate 
a single Arabic sound (e.g. sh, dh), many writers place a ligature 
under the letters (as sh, dh). It is rare, however, to find these 
combinations of letters indicating two Arabic sounds. Conse- 
quently it seems reasonable to use these pairs of letters without 
ligature for the single sound, and to find some other way of mark- 
ing the cases where they represent two sounds. For this I suggest 
the apostrophe. This could not be confused with hamzah by the 
Arabist, and it would indicate to the non-Arabist that the letters 
did not coalesce. This apostrophe is only absolutely necessary in 
cases where neither letter has a dot (of which there are none in 
this book), but I have used it where there was a dot or dots, and 
even in a word like * Ash'haP. I hope this innovation may commend 
itself to fellow orientalists. 

With regard to the form of Arabic names also I have tried to 
avoid puzzling the non-specialist. A brief explanatiqn here may be 
of value, however. An Arab's name has several parts. Thus 
Muhammad could be called Abu Qasim Muhammad b. ' Abdallah 
al-Hashimi al-Qurashi, that is, the father of Qasim, Muhammad, 
son of 'Abdallah, of (the clan of) Hashim, of (the tribe of) Quraysh. 
Any part of this name that is sufficiently distinctive may be used 
by itself. With a few exceptions (such as Ibn Ubayy for 'Abdallah 
b. Ubayy) I have kept to one form for each man. The last part of 
the name is often a nisbah or relative adjective, formed by adding T, 
and usually indicating at this period the tribe or clan to which 
a man belonged. * 


Both Christian and Muslim dates have generally been given, 
the Muslim months being indicated by Roman numerals since the 
names would convey little to most readers. Muslim dating is 
convenient when dealing with the sources, but the Christian dating 
is essential in order to understand the relation of Muhammad's 
career to Byzantine and Persian history. 

The Qur'anic quotations are normally from Richard Bell's 
Translation, by kind permission of the publishers, Messrs. T. & 
T. Clark. For help of various kinds my thanks are due to the 
Reverend E. F. F. Bishop and Glasgow University Library, to 
Professor J. Robson, to Professor G. H. Bousquet, to Mr. J. R. 
Walsh and to Dr. Pierre Cachia. For the compilation of the index 
and other secretarial assistance I am greatly indebted to Miss 
Elizabeth Whitelaw. 

W. M. W. 





1. The Situation at the Hijrah i 

2. The Earliest Expeditions 2 

3. The First Fighting 5 

4. The Battle of Badr 10 

5. The Situation after Badr 141 


1. Muhamrr^ad prepares for the impending Struggle 17 

2. Meccan Reactions to Badr 19 

3. The Battle of Uhud 21 

4. The Rousing of the Nomads 29 

5. The Siege of Medina 35 


1 . The Expeditions of the Year after the Siege 40 

2. The Expedition and Treaty of Al-Hudaybiyah 46 

3. After Al-Hudaybiyah 52* 

4. Meccan Reactions to Muhammad's Successes 55 

5. The Submission of Mecca 65 

6. The Battle of Hunayn 70 

7. The Consolidation of Victory 73 


1. The Tribal System confronting Muhammad 78 

2. The Tribes to the West of Medina and Mecca 82 

3. The Tribes to the East of Medina and Mecca 87 

4. The Tribes to the North 105 

5. The Tribes to the South of Mecca 117 

6. The Tribes in the Rest of Arabia 130 

7. The Success of Muhammad's Policy 142 


1. Social and Political Groupings before Muhammad 151 

2. Muhammad's Supporters 174 

3. The Muslim Opposition 180 



1. The Jews of Yathrib 192 

2. The Jews at the Hijrah 195 
3? Muhammad's Attempts to Reconcile the Jews 198 

4. The Intellectual Attack on the Jews 204 

5. The Physical Attack on the Jews 208 

6. Conclusion 219 


1. The Constitution of Medina 221 

2. The Position of Muhammad 228 

3. The Character of the Ummah 238 

4. Finance 250 


1. Security of Life and Property 261 

2. Marriage and the Family 272 

3. Inheritance 289 

4. Miscellaneous Reforms 293 

5. Conclusion 3^0 


1. The Religious Institutions of Islam 303 

2. Islam and Arab Paganism 309 

3. Islam and Christianity 315 


1. Appearance and Manner 321 

2. The Alleged Moral Failures 324 

3. The Foundations of Greatness 334 


A. Further Remarks on the Sources 336 

B. List of Expeditions and Dates 339 

C. Slaves and Freedmen among the Emigrants at Badr 344 

D. Muhammad's Letters to the Princes 345 

E. 'Those whose hearts are reconciled* 348 

F. Text of Selected Treaties 354 

G. The Treaties with Dumat al-Jandal 362 
H. List of Administrators sent out by Muhammad 366 
I. Zakdt and adaqah 369 


J. Marriage and the Family in pre-Islamic Times 373 

K. The technical terms in Surahs 4. 24/28, 5. 5/7, and 24. 33 389 

L. Muhammad's Marriages 393 

INDEX 401 


Aghdni = Abu '1-Faraj al-Isfahani, Kitdb al- Aghdni (20 vols.), Bulaq 

Bell, Translation = Richard Bell, The Qur'dn y translated with a critical 
rearrangement of the surahs (2 vols.)> Edinburgh, 1937-9. (See also 
under Q.) 

al-Bukhari al-Bukhari, Al-Jdmi as-Sahfy. The reference is by the name 
(and number) of the Kitdb and the number of the Bab. 

Buhl, Muhammad = Frants Buhl, Das Leben Muhammeds (German tr. by 
H. H. Schaeder), Leipzig, 1930. 

Caetani = Leone Caetani, Annali delV Islam, Milan, 1905, &c. 

El (i) = Encyclopaedia of Idam (4 vols. and supplement), Leiden, 1913, 

El (2) Encyclopaedia of Islam (second edition), Leiden, 1954, &c. 
El (S) = The Shorter Encyclopaedia of Islam, Leiden, 1953. 

IH = Ibn Hisham, Kitdb Sirat Rasul Allah (Das Leben Muhammeds 
nach . . . Ibn Ishdk bearbeitet von . . . Ibn Hischdm), ed. F. Wiistenfeld 
(2 vols.), Gottingen, 1859-60. 

IS = Ibn Sa'd, Tabaqdt (Ibn Saad, Biographien . . .), ed. E. Sachau 
(9 vols.), Leiden, 1905, &c. 

MJ Mecca W. Montgomery Watt, Muhammad at Mecca, Oxford, 1953. 

Q = Qur'dn (with the official Egyptian numbering of the verses, followed 
by Flugel's where it differs) ; the translation is normally that of Richard 
Bell, but occasionally, for the sake of harmonizing with the context, 
this has been modified or replaced by some other rendering. In 
certain places an attempt has been made to indicate by letters Bell's 
provisional dating of passages, as follows : 

A = very early. 

B = early, $rly Meccan, early Qur'an period. 

c = Meccan. 

D late Meccan. 

E = early Medinan. 

E-f = Medinan. 

F = connected with Badr. 

G = connected with Uhud. 

H = up to al-Hudaybiyah. 

i = after al-Hudaybiyah. 

= revised. 

Tab. = at-Tabari, Ta'rtkh ar-Rusul wa 'l-Muluk ('Annales'), ed. M. de 
Goeje (15 vols.), Leiden, 1879- 190 it All references are to the page 
of the Prima Series; the third volume of this contains the life of 


Muhammad up to A.H. 8 (pp. 1073-1686); the rest is in the fourth 

Tab., Tafsir = at-Tabari, Jam? al-Bayanfi Tafsir al-Qur'dn (30 vols.), 
Cairo, (1903)71321. 

Usd = Ibn al-Athir, Usd al-Ghabah (5 vols.), Cairo, (1869)71286. 
WK = al-Waqidi, Kitab al-Maghazi t ed. von Kremer, Calcutta, 1856. 

WW = al-Waqidi, Kitab al-Maghazt> tr. J. Wellhausen (Muhammed in 
Medina; das ist Vakidi's Kitab alMaghazi in verkiirzter deutscher 
Wiedergabe), Berlin, 1882. 

Particulars of other works are given at the first occurrence, and this 
may be found in the Index under the author's name. 




THE Medinan period of Muhammad's career begins with his 
arrival at Quba' in the oasis of Medina on or about 4 Sep- 
tember 622 (i2/iii/i). Life in Mecca had become intolerable 
or even impossible for him, owing to the opposition he had 
aroused, and he had come to an agreement with the leading men 
of Medina. The precise nature of this agreement will be discussed 
later. On the religious side it meant the acceptance of Muhammad 
as prophet, and on the political side the acceptance of him as arbiter 
between the opposing factions in Medina. Many seem to have been 
sincere in their acceptance of his prophethood, but others probably 
looked only at the political side. Relying on this agreement, some 
seventy of Muhammad's Meccan followers preceded him to 
Medina, where they were given lodging by his Medinan adherents. 
Thus on Ms arrival at Medina Muhammad had a large religious 
following and a position in the community of some political 
importance, though his powers may not have been exactly defined. 
His Meccan and Medinan followers came to be known respectively 
as the Emigrants (mul\ajirun y those making the hijrah) and the 
Ansar or 'helpers 1 . 1 

After a few days at Quba' Muhammad rode towards the centre 
of the oasis, and selected a spot for his house, being lodged near by 
until it was built. This house and courtyard later became the 
mosque, which is at the centre of the modern town ; but it is doubt- 
ful whether in Muhammad's time the population was denser here 
than at other parts of the oasis. The earliest settlements were more 
to the south in the 'Aliyah or 'high lands'. Here then Muhammad 
settled in the midst of his followers. During the first months 
at Medina he must have been busy directing and ordering the 
affairs of his community, religious and secular. His activity had 
many sides, but, except in the case of the external affairs the 
'expeditions' there are few chronological data. It is therefore 

1 The name of Ansar is probably derived from Q. 61 . 14; cf. 3. 52/45 ; 8. 72/73 ; 
9. loo/ioi, 117/118. Cf. also J. Horovitz, Koranische Untersuchungen, Berlin, 
1926, 99 f., with further references. 

6788 B 


convenient to deal first with external affairs, that is, relations with 
the pagan Meccans and with the nomadic tribes, and then to con- 
sider, in systematic rather than chronological order, the various 
internal aspects of the life of the Muslim community. 


The first attempts to collect biographical material about Mu- 
hammad were called al-maghazt, that is, the 'expeditions' or 'cam- 
paigns' ; and, although the Medinan period was not entirely filled 
with expeditions, these did play a large part in it, and it is natural 
to date an event roughly by its relation to some expedition. Of 
the seventy-four expeditions listed by al-Waqidi seven are assigned 
to the first eighteen months after the Hijrah. 1 They are of slight 
importance, in that nothing seemed to happen, but they are excel- 
lent illustrations of Muhammad's attitude towards the Meccans 
shortly after his departure from their city. 

The chief point to notice is that the Muslims took the offensive. 
With one exception these seven expeditions were directed against 
Meccan caravans. The geographical situation lent itself to this. 
Caravans from Mecca to Syria had to pass between Medina and 
the coast. Even if they kept as close to the Red Sea as possible, 
they had to pass within about 80 miles of Medina, and, while at 
this distance from the enemy base, would be twice as far from their 
own base. The attackers thus required to deal only with the force 
accompanying the caravan, and would easily be home before any 
rescue party came near them. The idea in these expeditions, as in 
most of the fighting of the desert Arabs, was doubtless to catch 
the opponents at a disadvantage by ambushing them, for in- 
stance. In these early expeditions the favourable opportunity 
apparently did not present itself; but its absence did not make the 
Muslims sufficiently desperate to risk a frontal attack. They merely 

In the first two or three expeditions the numbers involved are 
given as from 20 to 80. In those of the later part of 623 (ii-vi/2), 
however, when Muhammad himself took part, they are said to 
have ranged up to 200. The caravans attacked were mostly large 
one of 2,500 camels is mentioned and figures of 200 to 300 are 
given for the men accompanying them. These numbers are perhaps 

1 Details will be found in Excurrus B, pp. 339 ff. below; references given in 
this Excursus will frequently be omitted from the footnotes. 


exaggerated, since there were only 70 men with the important 
caravan which was the occasion of the battle of Badr, but it was 
conceivably its weakness which made Quraysh come out in force. 

If Muhammad had with him as many as 200 men, or even 150, 
the Ansar must have taken part. The sources tend to agree that 
the great expedition of Badr was the first on which Ansar were 
present; but they are not quite unanimous. In the only passage 
where the authorities are named 1 suspicion is roused by the fact 
that the last names in the chain of authorities are already late and 
are the names of persons much involved in legal disputes; in addi- 
tion all belong to B. Makhzum of Quraysh. Since Muhammad had 
well under a hundred Emigrants with him at Badr, where practi- 
cally all were present, there is a strong presumption (unless the 
figures are completely wrong) that the Ansar took part at least in 
the larger of the Sarly expeditions. 

Muhammad had indeed some opportunities for getting men in 
addition to the original Emigrants. Further Meccans joined him, 
'Ayyash b. Abl Rabl'ah (Makhzum) and Hisham b. al-'As (Sahm) 
are said to have left Mecca in the course of A.H. i, their departure 
perhaps being connected with the death of al-Walld b. al-Mugh- 
irah, the old chief of Makhzum. Miqdad b. c Amr and 'Utbah b. 
Ghazwan (confederates of B. Zuhrah and B. Nawfal respectively) 
changed sides during the expedition of 'Ubaydah. Further, some 
nomads may have been attracted to him for material reasons; this 
certainly happened in 'later years, but may not have occurred 
before Badr. These additional sources, however, could not bring 
the numbers up to 150. Apart from this argument about numbers, 
the action of Majdi b. *Amr of B. Juhaynah, during the expedition 
of Hamzah, in mediating between the Muslims and a superior 
force of Meccans, is probably due to the presence in the raiding 
party of Medinans, confederate with Juhaynah. 2 

Although there was no fighting on any of these seven expedi- 
tions, they were not without positive result for the Muslims. The 
mere fact that Muhammad was able to go about with a compara- 
tively large force, and appeared to have the intention of attacking 
the redoubtable Meccans, must have impressed the nomads. 
B. Juhaynah was presumably already friendly, since some at least 
were confederates of some of the Ansar. In addition, pacts of mutual 
non-aggression were made with Banu Damrah and B. Mudlij. 

1 WW, 33, last part of i. * * IH, 419; WW, 33. 


According to one account the terms were that neither party was 
to make raids on the other, to join in hostile concentrations against 
the other, or to help the other's enemies. 1 

In all this we may see a deliberate intention on Muhammad's 
part to provoke the Meccans. In so far as the Ansar joined in they 
must have been aware of the plan; indeed, their leaders were 
presumably aware of Muhammad's policy before they invited him 
to Medina. It is difficult, however, to know how far ahead Mu- 
hammad was looking at this moment. Was his aim primarily 
negative, to destroy the trade of Quraysh ? Or did he look beyond 
this to a conquest of Mecca ? There are no signs at this period that 
Muhammad had any thoughts of securing the Meccan trade for 
Medina (though one of the later expeditions was partly com- 
mercial); he still was not sufficiently strong in Medina and did 
not have any manpower to spare. The o'ne early expedition which 
was not against the Meccans, that against Kurz al-Fihri, illustrates 
the dangers against which he had to be constantly on guard ; it was 
an attempt to punish a freebooter of the neighbouring region for 
stealing some of the Medinan pasturing camels. 

Though Quraysh suffered no losses they were probably seriously 
perturbed at the threat to their trade. Despite the fullest precau- 
tions on their part the chances were that one day the Muslims 
would find the opportunity they looked for, and that would mean 
serious loss to Quraysh. For the moment Quraysh did nothing, 
but their eagerness to fight the Muslims at Badr is a measure of 
their annoyance. 

The Qur'an does not refer explicitly to the early expeditions, 
but it gives some glimpses of the attitude of the Muslims to 
fighting. What appears to be the earliest passage implies that the 
Emigrants wanted to fight, since it speaks of God permitting them 
to do so ;f permission is granted to those who fight because they 
have suffered wrong, . . . who have been expelled from their dwell- 
ings without any cause (or justification) except that they say, "Our 
Lord is God" '. 2 ^ater the Muslims, presumably both Emigrants 
and Ansar, receive by revelation a direct command to fight; * fight 
in the way of God, and know that God is one who hears and 
knows'. 3 There was apparently, however, some disinclination to 

1 IS, ii/i. 3. 17-20; cf. p. 84 below and the translation on p. 354. 

2 22. 39/40 B-. c 

3 2. 244/245 'early Medinan' (), but perhaps after Badr; cf. 9. 123/124. 


obey this command, for there are several references to the un- 
willingness of many to fight. 1 A fresh incentive is therefore given; 
the Muslims are told that God prefers fighters to those who sit 
still, that is, remain inactive at home; for the fighters there is 
a 'mighty hire', a reward in Paradise. 2 Clearly the Muslims re- 
garded their political and military activities as taking place within 
a religious setting. 


The standard account of the first fighting between the Muslim 
Emigrants and the pagan Quraysh, that during the expedition to 
Nakhlah, is the one in the version of Ibn Ishaq, which was based 
on the report of 'Urwah and transmitted by az-Zuhri and also 
Yazid b. Ruman^ According to this account 'Abdallah b. Jahsh, 
a confederate of B. 'Abd Shams, was sent out with a small party 
of from eight to twelve, all Emigrants. Muhammad gave him 
a sealed letter of instructions which he was not to read till they 
were two days' journey from Medina. In due course they opened 
the letter, and found in it an order to proceed to Nakhlah on 
the road from at-Ta'if to Mecca and there to ambush a Me9can 
caravan. They eventually met a Meccan caravan at Nakhlah and 
lulled the suspicions of the Meccans by making out that they were 
pilgrims. Having found a suitable opportunity they attacked. Of 
the four Meccans who ^appear to have been the sole attendants of 
the caravan, they killed one, 'Amr b. al-Hadrami, and captured 
two, while the fourth escaped. Although warning was thus given 
to the enemy, the party appears to have had no difficulty in bring- 
ing the caravan and their prisoners back to Medina. There, how- 
ever, some misgivings were expressed on account of the Meccan 
having been kiUed in the sacred month of Rajab, when bloodshed 
was forbidden. Muhammad at first kept the booty undistributed 
and did not accept the fifth they offered him. But eventually a 
revelation 4 justified their action. A Meccan deputation came to 
Medina to arrange for the ransom of the prisoners, and Muhammad 
agreed to do this for 1,600 dirhams apiece after the safe return of 
two Muslims who had become separated from the rest of the party. 

The first point to notice in this story is that Muhammad took 

1 E.g. 2. 216/212 E; 4. 77/79 ? E; 2. 246/247. 

2 4- 95/97 B-, 77/79 ? E; 3. 195/194- * 

IH, 423-7; WW, 34-37- 4 2- 217/314- 


elaborate precautions to ensure secrecy. Not merely did he give 
the leader of the expedition sealed instructions, whose content 
was presumably known only to himself, his scribe, and one or two 
trusted advisers; he also sent the party off by the Najd road, 
roughly in an easterly direction, although their ultimate goal was 
almost due south. Doubtless this was all done to prevent the 
Meccan espionage system from discovering what he was about. 
Perhaps some of the previous expeditions had failed through their 
intentions being communicated to the enemy. But in any case, as 
this expedition was to go to a spot much closer to Mecca than to 
Medina, it would have been dangerous for the participants had 
any news of their plan leaked out. 

Further, the reason for the hesitation of the party when they 
read Muhammad's orders was almost^ obvious danger 
of the enterprise and not any scruples about possibly dishonour- 
able aspects of what they were asked to do. The Arab may be 
recklessly daring when his blood is up, but in cold blood he tries 
to avoid serious danger. That was doubtless why Muhammad told 
'Abdallah b. Jahsh to send back anyone who was not entirely 
willing to carry out the plan (if this part of the source material is to 
be accepted). In this connexion the case of Sa'd b. Abl Waqqas 
and the companion who shared his camel, 'Utbah b. Ghazwan, is 
of interest. That their camel had strayed and that in their search 
they became cut off from the main party jvas the story they told 
when they got back to Medina several days after the successful 
raiders. But, while it may be a fact that this was the story they 
told, it does not follow that the story is true. One version suggests 
that it is not. 1 The two had certainly wasted a lot of time, and it is 
curious that this should be in the territory of B. Sulaym, the tribe 
of 'Utbah's birth. Another unfortunate incident -at a later date 
also tended to mar Sa'd's reputation for courage. At the great 
battle of Qadisiyah in 635/14 which broke the power of the Persian 
empire Sa'd commanded the Muslims, but owing to illness had 
to direct his forces from a litter at the rear. It is probably to 
counteract the bad appearance of these two incidents that so much 
is made in the traditional material of the fact that Sa'd was the 
first to strike a blow for Islam. 2 Much more is made of this than 
of the fact that, in killing 'Amr b. al-Hadrami at Nakhlah, Waqid 
b. 'Abdallah was the first t6,kill a man for the cause of Islam. 

1 Tab. i. 1278. 19-1279. i. * IH, 166; IS, iii/i. 99 f-J &c. 


Probably this difference is due to the fact that Waqid died at 
the beginning of the caliphate of *Umar and left no descendants, 
whereas Sa'd lived for a further forty years or so and became one 
of the* leading men in the state with a numerous progeny. Many 
of the notices of Sa'd's feat (which contain several discrepancies) 
are from himself or members of his family. 

The essential part of Muhammad's sealed orders to 'Abdallah 
b. Jahsh was to go to Nakhlah and ambush a caravan of Quraysh. 
The further clause (in some versions) about bringing back a report 
to Muhammad is clearly a later addition intended to give the word 
tarassadU the meaning 'keep a watch* instead of 'lay an ambush'; 
in this way all responsibility for blood shedding would be re- 
moved from Muhammad. There can be no doubt, however, that 
Muhammad sent out the raiders on an errand which he realized 
might involve dSaths among both his own men and the enemy. 
It is not clear whether Muhammad knew definitely that this par- 
ticular caravan would be passing Nakhlah about this date, or 
whether he sent the men because of the general probability that 
there would be caravans then. He may have surmised that any 
caravans on the comparatively safe route from at-Ta'if to Mecca 
would be lightly guarded in view of the efforts that had been 
expended on protecting the caravans to Syria which had to run 
the gauntlet of Medina. It seems most likely that Muhammad 
acted on the general probability, but it is not impossible that he 
had specific information. 

A more serious question is whether Muhammad expected the 
caravan to be ambushed during the sacred month of Rajab. Al- 
Waqidi places the incident at the end of Rajab and indicates that 
the Muslims, if they were to attack this caravan at all, had either 
to do so during the sacred month or else after it had entered the 
sacred territory of Mecca. If this account could be accepted, it 
might be the case that the caravan had upset Muhammad's 
calculations by being a little before time. The suggestion that the 
attackers were uncertain whether the sacred month had or had not 
ended looks like an attempt to whitewash what is known to be 
black. It is very suspicious that, while some sources say the date 
was the end of Rajab, others make it the beginning. It seems that 
it was only known that the event was alleged to have taken place 
in Rajab, and that the rest is extenuating conjecture. If that is so, 
the incident may well have taken f>lace about the middle of the 


month, and the orders given by Muhammad may have contem- 
plated an ambush during the sacred month. 

If we suppose that Muhammad intended the violation of the 
sacred month (although it is by no means proved that he did so), it 
does not mean that he was contemplating anything scandalous or 
dishonourable. The sacredness of the month of Rajab was bound 
up with the pagan religion which he was denouncing. Violation of 
the sacred month would be on a par with the destruction of idols. 
But, on this supposition, what are we to make of Muhammad's 
hesitation before accepting a fifth of the booty ? We cannot fairly 
regard him as abandoning his companions, nor as being afraid of 
Quraysh although some of the Jews of Medina made puns on the 
names of slayer and slain which indicated that war was as good as 
declared. The easiest solution is to hold that after the event he 
discovered that there was a far stronger .feeling ofo the question of 
violation than he had anticipated. Possibly many were afraid of 
the punishment to be meted out by offended deities, a punishment 
which might affect the whole community if they accepted the 
booty. Others certainly pointed to the contradiction between this 
breaking of Divine Law and Muhammad's call to, worship and 
serve God; 'Muhammad imagines that he is keeping obedience to 
God, they said, but he is the first who has profaned the sacred 
month and he has killed our comrade during Rajab'. 1 

The question of the sacred months is a difficult one. 2 Muslim 
writers, following the Qur'an, 9. 36, hold that four always had 
been regarded as sacred, namely, Rajab, Dhu '1-Qa'dah, Dhu 
'1-Hijjah, al-Muharram (vii, xi, xii, i, respectively). But elsewhere 
2. 194/190; 2. 217/214; 5. 2; 5. 97/98 the Qur'an speaks of 
'the' sacred month. It has been suggested that different districts 
had different usages, and that the number four is an attempt at 
compromise. This may explain why the reaction in Medina to the 
incident at Nakhlah took Muhammad by surprise. Perhaps also 
the Medinans clung more closely to the old beliefs than the 
Meccans, for the latter had had experience of a war in which 
sacred things were violated. 

The revelation which ended Muhammad's hesitation ran thus : 3 

They will ask thee about the sacred month, fighting therein; say: 

1 Tab. 1278, 3, from as-Suddi. For the 'fifth* cf. p. 255 below. 
* Cf. M. Plessner, arts. 'al-!NJutiarram' and 'Radjab' in El (i). 
3 Q. 2. 217/214. 


'Fighting therein is serious, but debarring [people] from the way of 
God, and unbelief in Him ... is in God's sight more serious still* ; 
persecution is more serious than killing. . . . 

This admits that the violation of the sacred month was 'serious', 
but it reminds the Muslims that the offences of Quraysh against 
God were more serious; the intended inference is perhaps that 
punishment is more likely to fall on Quraysh. The word 'serious' 
(kablr) might almost have the connotation of 'sin', but the verse 
is not a prohibition of fighting for the future (though some Muslim 
writers take it in this way and then say it is abrogated) ; it is rather 
a justification of what has been done in the past. It may be that 
for some years after the affair at Nakhlah Muhammad tried to 
avoid giving offence by not sending out expeditions in Rajab. 
Al-Waqidi indeed mentions two in that month, in the years 6 and 8 
(627 and 629), but on neither apparently was any enemy blood 
shed though some Muslims were killed in 6. 1 Several expeditions 
undoubtedly took place in the other sacred months; and in the 
light of this and of other matters it is commonly stated by Muslim 
scholars that the prohibition of fighting in the sacred months was 
abrogated. 2 

Among the probabilities and uncertainties through which we 
have been wading there is a little firm ground. It is clear that 
people accused 'Abdallah and his party of having violated the 
month of Rajab, and also that Muhammad was not in a position to 
demonstrate that the Muslims had not done so. It is tolerably 
certain that Muhammad himself had few scruples about fighting 
in the sacred months, but that he had to respect the scruples of an 
important section of his followers and to guard against repercus- 
sions which might weaken his prophetic authority. And it must 
be insisted that, even if Muhammad intended the raiding party to 
violate the sacred month, there was in Arab eyes nothing dis- 
honourable or disgraceful about that, especially in view of his 
general attack on paganism. Finally, in addition to losing a life 
and a valuable caravan Quraysh would be thoroughly infuriated 
by the fact that these acts were perpetrated under their very noses, 
as it were. 

1 See Excursus B. 

* Cf. WW, 29-31; Tab., Tafsir, ii. i9<\(on Q. 2. 217/2x4); AbQ Ja'far an- 
, K. an-Ndsikh wa 'l-Mansukh, Cairo, 1938/1357, 32 f. 



(March 624 = ix/2) 1 

The booty from Nakhlah gave a fillip to the policy of raiding 
Meccan caravans, and for his next expedition Muhammad was 
able to collect 300 men, at least a hundred more than on any 
previous occasion. The excess was doubtless mainly from the 
Ansar, since it may be assumed that practically all the Emigrants 
took part in the earlier large raid. According to Ibn Sa'd's reckon- 
ing there were 238 of the Ansar at the battle of Badr and only 86 
Emigrants. 2 Muhammad apparently heard in good time that a large 
caravan was setting out from Gaza to return to Mecca and realized 
that it was well worth plundering. Although only 70 men (or 
perhaps even fewer) accompanied it, the merchandise was later 
said to be worth 50,000 dinars. All the leading Meccan merchants 
and financiers had an interest in it; indeed, nearly everyone 
in Mecca was concerned for its safe return. Perhaps several 
smaller caravans some of them having been the object of Muslim 
attentions on their way north had joined together for greater 
safety. 3 

In charge of the caravan was Abu Sufyan b. Harb, one of the 
most astute men in Mecca. He seems to have realized at an early 
stage that Muhammad would try to attack his caravan and to have 
sent a timely request to the Meccans to dispatch a force to cover 
the caravan at the danger-point. (Some sources say that he sent 
his message only after hearing of Muhammad's preparations, but 
considerations of timing make this impossible.) 

The Meccans, led by Abu Jahl, responded to Abu Sufyan's 
message by sending a large force, said to be about 950. Nearly all 
the fighting men of Mecca went, after a neighbouring chief of the 
B. Kinanah had given his word that, even if Mecca were denuded 
of defenders, it would not be attacked by the section of Kinanah 
which had a blood-feud with Quraysh. The size of the force shows 
that Abu Jahl probably intended to overawe Muhammad and his 

1 IH, 427-539; WW, 37-90; Tab. 1281-1359; Caetani, i. 472-518; M. 
Hamidullah, The Battlefields of the Prophet Muhammad, Woking, 1953/1373, 
11-17 (reprinted, with continuous paging, from The Islamic Review, 1952, 1953). 
The strategy of this and other battles is discussed by Muhammad 'Abd al- 
FattSb Ibrahim in Muhammad al-Qd'id, Cairo, 1945/1364. 

a IS, iii/2, iii/i. 

3 IH, 421 f., 428; WW, 34, 39 f.; contrast Caetani, i. 463, n. i. 


followers and any potential followers, and so to scare them from 
meeting him in battle and from raiding caravans in the future. Some 
days out from Mecca Quraysh got word that the caravan had eluded 
Muhammad and was safe. The only cause of war now was the blood 
of 'Amr b. al-Hadrami, and 'Utbah b. Rabi'ah of 'Abd Shams was 
ready to pay blood-money to keep peace, but Abu Jahl skilfully 
shamed 'Utbah into withdrawing his offer, and so forced Quraysh 
to advance ; he was presumably hoping to get rid of Muhammad 
once for all. 

Such, at least, is the account of the Meccan proceedings given 
by several sources. Even if they are unduly hostile to Makhzum 
and friendly to f Abd Shams, there is probably much truth in what 
they say. There was certainly no strong bond of unity among 
Quraysh; two c^n, Zuhrah and 'Adi, withdrew completely after 
it was decided, even though the caravan was safe, to advance to 
Badr. Doubtless they felt that Abu Jahl and his friends would 
stand to benefit most by the destruction of Muhammad. Fear of 
a conflict would have little to do with the action of the Meccans, 
since the majority presumably believed that the Muslims would 
not venture to attack them. 

The Muslims certainly did not expect a conflict when they set 
out. According to the oldest source, 'Urwah's letter to 'Abd al- 
Malik, 1 'neither the Messenger of God nor his Companions heard 
of the expedition of Quraysh until the prophet came to Badr'. Had 
the Muslims known there was likely to be a battle they might have 
shrunk from taking part in the expedition. There is a curious story 
of how some of Muhammad's party captured one of the Meccan 
water-carriers and questioned him; when he told them the truth 
about Abu JahPs force, they thought he was lying and punished 
him, but when he told them lies about Abu Sufyan they believed 
him, and it was only when Muhammad himself interviewed him 
that the real state of affairs was discovered. Whether in this way 
or some other way, Muhammad appears to have had definite news 
of Quraysh before they had any exact information about him, and 
so to have had the tactical initiative. The phrase in the Qur'an 
(8. 7) about God 'promising that one of the two parties (sc. the 
caravan or the relief force) should be yours* would seem to imply 
that Muhammad knew about Abu Jahl sufficiently long before the 
battle for it to be uncertain with yvhich party contact would be 
1 Tab. 1284 ff.; tr. by Caetani, i. 472 ff. 


made. It is also said that the Ansar were pledged to defend 
Muhammad only within Medinan territory and that, before com- 
mitting himself to a course leading to battle, Muhammad conferred 
with them and asked if they would support him in these circum- 
stances. It is conceivable that when the Muslims learnt about 
Quraysh they were so close to them that retreat would have in- 
volved loss of face ; but it is more likely that Muhammad saw an 
opportunity of attacking Quraysh with conditions in his favour, 
and managed to convince his followers of the soundness of such 
a course. 

The date given for the battle is the i7th, igth, or 2ist of Rama- 
dan (ix) in the year 2 ( = 13, 15, or 17 March 624). On the night 
before it Muhammad, aware that Abu Jahl was making for Badr, 
seized the water-supply there, blocked up all the wells except one 
round which he stationed his men the* one nearest to Mecca 
and so forced his opponents, presumably now in need of water, 
to fight on ground and under conditions of his choosing. Quraysh 
was not ambushed, but it was apparently placed in a position in 
which it could not avoid fighting, though the conditions were 
unfavourable. If the sources can be trusted on poirlts of detail, 
Quraysh on the evening before the battle learnt that Muhammad 
was close to them, but not his precise whereabouts. Next morning 
his presence at the wells took them by surprise. Nevertheless there 
seems to have been a series of single combats between champions, 
the normal prelude to Arab set battles. There was also arrow- 
shooting on both sides, and latterly a general me!6e, which turned 
into the flight of Quraysh. In the course of the battle from forty-five 
to seventy of Quraysh were killed, including Abu Jahl himself and 
several other leaders. A similar number were taken prisoner. For 
the Muslims there was much booty, and to prevenjt the quest for 
loot from interfering with the pursuit of the enemy, Muhammad 
had to announce that the booty, apart from the spoils from those 
killed and the ransoms of those taken prisoner, would be divided 
equally among those who took part in the battle. 

One or two of the prisoners were treated with the harshness and 
ferocity which were probably not unusual among the Arabs of 
that age. The common attitude was that a man might do what he 
liked with his prisoner; the only point for him to consider was 
what was profitable or advisable for himself and his clan. At Badr 
in at least one instance a pagan was being led off by a Muslim 


captor when a group of Emigrants who particularly hated him 
noticed him, and at once set on him and killed him; the captor 
incidentally lost the potential ransom. Such excesses Muhammad 
put a stop to. In general his policy was to hold prisoners to ransom, 
but those belonging to his own clan or in some other way specially 
related to the Muslims, and those not sufficiently influential or 
wealthy to be ransomed, he usually set free without ransom. He may 
already have begun to realize how important it was going to be for 
him to win the hearts of the Meccans. An exception to this lenient 
policy was f Uqbah b. Abl Mu'ayt, who was executed for his former 
hostility to Muhammad, and in particular because he had composed 
verses about him. An-Nadr b. al-Harith, who had claimed that his 
stories about things Persian were as good as those of the Qur'an, 
was likewise executed. 1 

A number of factors 'combined to bring about this notable 
victory for the Muslims. One was the lack of unity among Quraysh, 
which has been noticed above. By defections their number had 
been reduced far below the original 950, perhaps to 600 or 700; 
and of these many were not whole-hearted supporters of Abu 
Jahl's policy. They were also over-confident. Against such a foe 
the spirit of the Muslims would count for much. Their belief in 
a future life probably gave them greater courage in battle, and 
Muhammad's confidence inspired them with confidence. His 
generalship also won for them a tactical advantage. These seem 
to be the main reasons for the Muslim victory. There are no 
grounds for holding that the fighting qualities of the Ansar were 
superior to those of Quraysh. The list of pagans killed often 
mentions the name of the Muslim who killed a particular man, 
but too much reliance cannot be placed upon details; the large 
numbers kille4 by 'Ali and Hamzah must be exaggerations. If, 
however, after allowing for exaggeration, we may accept the general 
impression made by the list and other relevant details, it is that the 
farmers of Medina were not markedly superior fighters to the 
merchants of Mecca. It has to be remembered that the pagans, or 
at least many of those killed, were considerably older than the 
majority of the Emigrants, and were probably suffering from thirst. 
Though the numbers of the Emigrants were only about one-third 
of those of the Ansar, they appear to have taken their full share 
in the battle. 

1 IH, 4 5 8ff.;WW, 78 ff. 



The loss of trained men was a disaster of the first magnitude for 
Mecca. Besides Abu Jahl (of Makhzum) the following leaders were 
killed: 'Uqbah b. Abi Mu'ayt, 'Utbah b. Rabl'ah, and Shaybah b. 
Rabfah of the clan of 'Abd Shams; al-Harith b. 'Amir and Tu'aymah 
b. 'Adi of Nawfal; Zam'ah b. al-Aswad, Abu '1-Bakhtari, and 
Nawfal b. Khuwaylid of Asad; an-Nadr b. al-Harith of 'Abd 
ad-Dar; Munabbih b. al-Hajjaj and his brother Nubayh of Sahm; 
and Umayyah b. Khalaf of Jumah. 1 There can hardly have been 
left alive in Mecca a dozen men of similar ability and experience. 
Abu Sufyan, of course, was safe with the caravan and now became 
the most prominent man in the city. Suhayl b. 'Amr was a prisoner, 
but was ransomed, and Hakim b. Hizam and various others 
managed to escape from the battlefield. There were younger 
men, too, coming forward. Nevertheless the catastrophe was 

With this went no little loss of prestige, even though there was no 
immediate change in the political situation. The Arabs of the 
Hijaz realized that this battle did not mean that Muslim Medina 
had replaced Mecca as chief power in the area. Further tests of 
Muhammad's strength were required before everyone flocked to 
him from far and near. It was now clear, however, that Abu Jahl 
had been correct in rating very high the seriousness of the threat 
from Muhammad, even if he had made errors of judgement in 
other points. Muhammad had now, as it were, thrown down a 
gauntlet which the Meccans could not honourably refuse to pick 
up. He had effectively challenged them to a full-scale trial of 

Although neither party when they left home expected to fight, 
the outcome of the battle was no accident. The Meftcan belief that 
Muhammad would be so overawed that he would avoid them 
rested on a misappraisal of the relative strength and fighting 
quality of the two parties; and this in turn was doubtless due to 
over-confidence in their own powers and failure to realize how 
much these had been weakened by the malaise of the age for 
example, by the reliance on money and other vices attacked in the 
Qur'an. Muhammad, on the other hand, though he might not 
have deliberately organized an expedition to seek an engagement 

IH, 5o?ff.;WW, 8iff. 


with Quraysh, had shown himself disposed to attack them in 
circumstances favourable to the Muslims. When chance made it 
difficult for him to avoid fighting at Badr or perhaps we should 
say, hfelped him to persuade the Ansar to fight he kept his head 
and made the fullest possible use of the opportunity. Thus, though 
the fighting was unpremeditated, the result fairly reflected the 
relative strength of the two sides within the limits of a small 
engagement. Of course, only a small part of the total available 
forces of Quraysh had been involved. It therefore remained to be 
seen whether Muhammad could increase the manpower at his 
disposal sufficiently rapidly to be in a position to hold the larger 
army which Quraysh were sure to put in the field against him. 

In Medina itself the victory considerably strengthened Muham- 
mad's position, which had perhaps been deteriorating during the 
previous few month's when it looked as if he was unlikely to achieve 
anything. Usayd b. Hudayr, for example, one of the chief early 
converts, was not sufficiently enthusiastic in his support of Mu- 
hammad to take part in the expedition, though he made his excuses 
as soon as Muhammad returned victorious. 1 Muhammad further 
used the flush of victory to eliminate some weaknesses. Two 
persons who had written poems against him 'Asma* bint Marwan 
of Umayyah b. Zayd and Abu 'Afak of B. 'Amr b. 'Awf were 
killed by persons belonging to their own or related clans, but 
nothing was said and no blood-feud followed. 2 About the same 
time the Jewish tribe of B. Qaynuqa' was attacked after a trivial 
dispute had led to the death of a Muslim, was besieged for a fort- 
night, and, when they surrendered, sent away from Medina. 
Thereby Muhammad's chief rival, 'Abdallah b. Ubayy lost perhaps 
as many as 700 of his confederates. 3 

The most important result of the battle, however, was the 
deepening of the faith of Muhammad himself and his closest Com- 
panions in his prophetic vocation. After years of hardship and 
a measure of persecution, after the weary months at Medina when 
nothing seemed to be going right, .there came this astounding 
success. It was a vindication of the faith which had sustained them 
through disappointment. Very naturally they regarded it as 
miraculous, the work of God, as the Qur'an asserted (8. 17): 'Ye 
did not kill them, but God killed them, and when thou didst 

1 WW, 38, ?2;cf. p. 181 below. 

2 IH, 994-6; WW, 90-92. 3 Cf. pp. 181 f., 209 f. below. 


throw, it was not thou but God who threw. . . .'* Moreover, this 
disaster which had overtaken the pagans was the punishment which 
had been foretold in the Meccan revelations, 2 and thus Muham- 
mad's claim to prophethood was verified. 

So much is certain. It is further probable that the word furqdn, 
at least in some passages of the Qu'ran, is to be interpreted as 
Richard Bell suggested. 3 In 8. 41/42 'the day of the furqdn, the day 
the two parties met* must be the day of Badr ; zndfurqdn, in virtue 
of its connexion with the Syriac word purqdna, 'salvation', must 
mean something like 'deliverance from the judgement'. This being 
so the furqdn which was given to Moses 4 is doubtless his deliver- 
ance when he led his people out of Egypt, and Pharaoh and his 
hosts were overwhelmed. Similarly, Muhammad's furqdn will be 
the deliverance given at Badr when the Calamity came upon the 
Meccans. That was the 'sign' which confirmed his prophethood. 
Perhaps there is also a reference to the experience, analogous to 
the receiving of revelation, which Muhammad apparently had 
during the heat of the battle, and as a result of which he became 
assured that the Muslims had invincible Divine assistance. 5 

1 For the traditional story cf. p. 312 below. 2 Q. 8. 30-35, &c. 

3 The Origin of Islam in its Christian Environment, London, 1926, 118 ff. 

4 Q. 2. 53/So; 21. 48/49. 5 WW, 54; cf. Q. 8. 43/45 f. 




^FTER his success at Badr Muhammad must have realized that 
yLA he was committed to * total war* with the Meccans. Their 
JL \. prosperity depended to a great extent on their prestige, and 
in order to maintain their position they must in no uncertain 
fashion retrieve what they had lost at Badr, in addition to loosen- 
ing Muhammad's hold on their route to the north. From the 
Meccans, therefore,, Muhammad could expect nothing but an 
intensification of the struggle, so that he must clearly devote all 
his energies to strengthening himself and weakening his enemies. 

The mere news of Badr and the sight of the booty had brought 
an accession of strength to Muhammad. The people of Medina 
were much readier to join Muhammad's expeditions; for that to 
Dhu Amarr in September 624 (iii/3) he was able to muster 450. 
The friendly tribes between Medina and the sea were presumably 
more ready to help Muhammad openly ; at least Quraysh did not 
venture to send any caravan to Syria by that route in the summer 
of 624. When Quraysh marched north for the battle of Uhud men 
of Khuza'ah passed information to Muhammad. Pagan nomads in 
the neighbourhood of Medina were much readier to profess Islam ; 
mention is made not merely of one or two individuals but also of 
the whole tribe of Muharib, who followed their chief, Du'thur 
b. al-Harith. 2 

The increased forces at his disposal were employed by Muham- 
mad to create a healthy respect for the new Medinan state among 
nomadic tribes friendly with Mecca. When Abu Sufyan raided 
Medina to show that Mecca was not 'down and out', Muhammad 
gave a counter-display of his power by pursuing him with at least 
200 men. The strong tribes of Sulaym and Ghatafan, which later 
helped Quraysh at the siege of Medina, were raided in the expedi- 
tion of al-Kudr and a large number of camels driven off. A little 
later another raid with 300 men was made against Sulaym, and, 

1 General sources for i and 2: IH, 539-55, 994-6; WW, 73-76, 90-101. 
* IH, 544J WW, 99-101. 

6788 C 


though no booty was captured, the expedition was doubtless not 
without effect. 

In all Muhammad's planning he was fully aware of the impor- 
tance for him of what may be called the ideological aspect and of 
his pre-eminence there. It was always possible for him to win 
enemies over to his side by converting them ; one at least of the 
pagans who came to arrange for the ransom of prisoners after Badr 
was so impressed by some of the things he saw in Medina that he 
became a Muslim, though previously he had been plotting to kill 
Muhammad. Muhammad's decision (contrary to the views of some 
of his supporters) that in general the prisoners from Badr were to 
be held to ransom is not simply a mark of leniency of disposition 
and of the great need for improving the financial position of the 
Muslims; it is perhaps also the beginning of tjie Realization that, to 
achieve the distant aims he was beginning to see over the horizon, 
he required the administrative abilities of the Meccans, and that 
therefore his task must be not to destroy Quraysh but to win them 
for his cause. 

Out of the same awareness of the importance of the ideological 
aspect sprang events like the assassinations of 'Asma' bint Marwan 
and Abu 'Afak who had made verses criticizing Muhammad, and 
the expulsion from Medina of the Jewish tribe of Qaynuqa'. The 
assassination of Ka'b b. al-Ashraf a little later was similar. Ka'b 
was the son of an Arab from the distant, tribe of Tayyi', but was 
reckoned as belonging to his Jewish mother's tribe of an-Nadir, 
in which he was one of the leading men. When he heard the news 
of Badr, he set out for Mecca, and by his verses helped to rouse the 
Meccans to grief and anger and the desire for revenge. Eventually 
he returned to Medina apparently because the Muslim poet 
Hassan b. Thabit ridiculed the families whose guest he was. 
Muhammad now made it known that he would welcome Ka'b's 
removal, and five doughty Muslims laid a plot. Abu Na'ilah, Ka'b's 
foster-brother, won his confidence by complaining about the state 
of affairs in Medina as a result of Muhammad's presence. The 
points mentioned are the general enmity of the Arabs, the difficulty 
of travelling, the payment of contributions (sadaqah), with the 
consequent impoverishment of their families and lack of food; 
these may be the points most keenly felt by Muhammad's oppo- 
nents at this period. Ka'b agreed to give the men food and was to 
receive their arms as security. This gave the conspirators an excuse 


for meeting Ka'b privately by night with their arms. He was 
overpowered and killed, and his head carried off and flung at 
Muhammad's feet. 

SucK measures made it clear that Muhammad was not a man to 
be trifled with. For those who accepted him as leader there were 
material advantages ; for those who opposed him there were serious 
disadvantages. Thus, apart from Muhammad's preaching, men 
had many reasons for taking his side. Consequently, although the 
Meccans had 3,000 men in the field at the battle of Uhud, the out- 
come of the battle is never said to be due to the fact that the Mus- 
lims were outnumbered; and, though this point could partly be 
explained away, it does perhaps give us some justification for 
thinking that Muhammad may have had about 2,000 men that day. 


The news of Badr was received at Mecca at first with incredulity, 
then with a dismay which inhibited all effective action. Abu Sufyan 
took control of affairs and for a time forbade mourning for the 
dead; this was ostensibly to prevent the Muslims gloating over 
their plight and to avoid dissipating the energies of the Meccans 
when all their strength was required to prepare for revenge, 1 but 
perhaps really to avoid a complete collapse of morale. For a similar 
reason he announced that he had vowed to have nothing to do 
with oil or women till he. had carried out a raid against Muhammad. 
The pent-up feelings, however, at length swept away the pro- 
hibition, which had perhaps served its turn. Ka'b b. al-Ashraf in 
his poems encouraged the expression of grief among the Meccans 
in order subsequently to stir up their desire for revenge. This and 
the need for restoring their position was no doubt uppermost in 
their minds once they had recovered from the first shock of the 
disaster. Thus Safwan b. Umayyah b. Khalaf of Jumah, whose 
father had been killed at Badr, persuaded a member of his clan, 
'Umayr b. Wahb by name, to try to assassinate Muhammad while 
in Medina negotiating for his brother's ransom (though what 
actually happened was that 'Umayr became a Muslim). A more 
positive step was Abu Sufyan's proposal to devote to preparations 
for war all the profits of the caravan he had brought safely back; 
those concerned seem to have agreed. 

Some ten weeks after Badr, Abu Sufyan, in fulfilment of his 
1 WW, 73 = WK, 114 f. 


vow, led a party of 200 (or 400) men to raid Medina. His primary 
aims were doubtless to restore confidence among the Meccans and 
to show the world that the day of Quraysh was not yet over. With 
such a force less numerous than the Muslims at Ba'dr he 
cannot have intended to inflict any serious damage on Muhammad, 
unless he expected more than half the inhabitants of Medina to 
join him (which is unlikely). He must have had a firm grip on all 
information leaving Mecca for he apparently reached the outskirts 
of Medina without Muhammad's knowledge. A friend, the chief 
of the Jewish tribe of the Nadir, gave him a meal and presumably 
information (if we may trust the account), but nothing more, and 
he decided to retreat immediately. To fulfil his oath two houses 
were burnt and some fields laid waste. On the way back the 
Meccans abandoned some excess provisions^ mainly barley-meal 
(sawlq), and this was picked up by the* Muslims. In consequence 
the expedition was known to the Muslims as 'the barley-meal raid'. 1 

In the existing state of affairs it was clearly wiser not to attempt 
to force the passage of a caravan to Syria through territory friendly 
to Muhammad, but to concentrate on raising which could 
destroy his power. No caravans were therefore sent to Syria by the 
usual route which passed between Medina and the sea. A group of 
Quraysh, however, headed by Safwan b. Umayyah (who was 
perhaps setting up as a rival to Abu Sufyan), decided to risk send- 
ing a caravan by a route well to the eas{ of Medina, and found 
a reliable guide. Unfortunately for them, however, Muhammad 
got wind of the plan, sent out Zayd b. Harithah with 100 men, and 
captured merchandise worth 100,000 dirhams; the men in charge 
of it escaped, doubtless being, at this period between Badr and 
Uhud, thoroughly terrified at the prospect of fighting with Muslims. 2 

Meanwhile a large force was being raised to go.against Medina. 
Embassies were sent to various tribes, including Thaqlf of at-Ta'if 
and the nomadic 'Abd Manat (which was closely related to Quraysh 
and included Bakr). The Ahabish who followed Quraysh were also 
summoned. 3 About n March 625 (25/ix/3) the Meccans set out 
with an army of 3,000 well-equipped men, of whom 700 had coats 
of mail; there were 3,000 camels and 200 horses. Abu Sufyan was 
in command, since to command in battle was one of the privileges 

1 IH, 543 ; WW, 94- 2 IH, 547 *'> WW, 100 f. 

3 WK, 199. 9, aoi foot; cf. al-WSljiidi, Asbdb an-Nusul, Cairo (i897)/i3i5, 
177; also p. 8 1 n. below. 


of his clan, but others, notably Safwan b. Umayyah, had an im- 
portant share in the direction of the campaign. The army advanced 
by easy stages and reached the oasis of Medina on Thursday, 
21 March (s/x). For their camp they selected a site on the far side 
of the oasis from Mecca near the hill of Uhud. In this neighbour- 
hood there were fields of corn, now in the ear, and they deliberately 
pastured their animals there in order to provoke the Medinans to 
come out to fight. Apart from this the peculiar Medinan strong- 
holds or atdm, at various points in the oasis, were able to hold out 
for far longer than the Meccans were capable of besieging them. 


(Saturday, 23 March 625 = 7/x/3) 

There is a great mass of material in the early sources about the 
battle of Uhud, 1 but much consists of accounts of trivial incidents 
redounding to the glory of individuals (and so preserved by their 
descendants or clansmen), or rebutting accusations against them. 
From this mass it is not possible to give a full or clear account of 
the battle. Nevertheless, if we accept the general soundness of the 
material, a rough outline does emerge. This runs somewhat as 

The Meccans advanced by Wadi 'l-'Aqiq and camped to the 
north of Medina near Uhud on Thursday, 21 March. Almost at 
once a scout brought Muhammad exact information about their 
strength, and some of the leading Ansar kept guard at Muham- 
mad's door all night. Early on the Friday a council of war was held 
in Medina. Muhammad, 'Abdallah b. Ubayy, and some of the 
senior men were for remaining in the centre where the buildings 
were closer, and so forcing the enemy to undertake a combination 
of siege and house-to-house fighting; but younger men, together 
with one or two men of weight, argued that to allow the Meccan 
army to lay waste their fields as it was doing would make them seem 
cowards and ruin their reputation in the eyes of the nomadic 
tribes; so they must go out to the enemy. Eventually Muhammad 
decided on this course, and, though some of the hotheads cooled 
down and said they were willing to accept Muhammad's original 
plan, he (very properly) stuck to his decision with the remark that 

1 IH, 555-638; WW, 101-48; Tab. 1383-1427; discussed in detail by Caetani, 
i. 540-66. Cf. Hamidullah, Battlefields of Muhammad, 18-24. See also the map, 
p. 152 below. 


once a prophet has put on armour he must not take it off until 
God has decided between him and his enemy. Later in the day 
the Medinan forces set out in the direction of the enemy camp. 
Muhammad is said to have rejected the help of a Jewish contingent, 
confederates of 'Abdallah b. Ubayy, because they were not be- 
lievers. Some distance short of the enemy they halted for the 
night, then very early next morning, using their superior knowledge 
of the terrain, they made their way unobserved to a position on the 
lower slopes of the hill of Uhud, with the enemy roughly between 
them and the city. The left flank was protected by fifty archers 
under 'Abdallah b. Jubayr. 

The departure of 'Abdallah b. Ubayy and his followers shortly 
before the battle is curious. The sources suggest that he was 
annoyed because Muhammad did not adopt ttye plan he advocated 
on the previous day. But this is difficult' to believe, since he seems 
to have gone with Muhammad right to the site of the battle. It is 
conceivable that he retired, in agreement with Muhammad, in order 
to defend the main settlement against a possible enemy attack. 1 In 
sources not friendly to him his motives could easily be blackened, 
especially when, after the battle, he made no secret of his joy at the 
discomfiture of his rival, Muhammad. If he acted independently 
it may have been from the calculation that neutrality during the 
battle would strengthen his position with both parties afterwards ; 
he is said to have acted in a similar way, at the battle of Bu'ath 
between the Aws and the Khazraj. 

The Meccans had to move forward to the attack across a wadi. 
The cavalry may have attacked first, but, if so, they were driven 
back by the Muslim archers. Then the Meccan standard-bearer 
moved forward, perhaps with a view to single combat, but soon 
a general melee developed round the standard. The clan of 'Abd 
ad-Dar who had the privilege of bearing the standard fought with 
great gallantry against overwhelming odds. Nine members of the 
clan seem to have been killed defending the standard a large 
number for a small clan. The standard did not fall into Muslim 
hands, but the Meccan forces withdrew before the Muslim on- 
slaught, perhaps even fled. However, as victory seemed to be 
almost within the grasp of the Muslims, there was a sudden reversal 
of fortune. The cavalry on the Meccan right under Khalid b. al- 
Walid, observing the Muslim ranks in some disorder and in 

1 Cf. WW, 138. 

ii. 3 THE BATTLE OF UtfUD 23 

particular the archers advancing from their post, quickly overran 
the few remaining archers and attacked the Muslim flank and rear. 
A scene of great confusion followed, especially as the cry went up 
that Muhammad had been killed. Muslims wounded other Mus- 
lims in at least one case mortally. Muhammad was not in fact 
killed, but for some time there was a fierce hand-to-hand struggle 
round him, in which he received two or three wounds on the face 
and leg, and inflicted a spear wound on one of the Meccans which 
caused the latter's death subsequently. 1 Eventually Muham- 
mad and the group round him managed to reach the slopes of 
Uhud, and here the Muslims were rallied and given some sort of 
order. A section, however, had become separated from the main 
body and had made for the stronghold of the clan of Harithah in 
the direction of the city; of these a number, perhaps nearly all, 
were killed. 2 Th positioi>on the hill the Muslims were perhaps 
a little higher up now than before the battle still had the advan- 
tages for defence which Muhammad saw when he originally chose 
it, and the Meccans soon ceased to attack, though they remained 
on the field for some time longer. With a final taunt to the Muslims 
Abu Sufyan ordered withdrawal, and both the fighting men and 
the baggage train moved off. For a time it seemed possible that 
they might attack the town of Medina itself, but they left it alone 
and headed for Mecca. 

Such then is the outline account of the battle into which the 
numerous incidents recorded by Ibn Ishaq and al-Waqidi are 
fitted. The unsatisfactory outcome of the battle is attributed to the 
disobedience of the majority of the archers. That this outline is in 
essentials the official Muslim account of the battle is confirmed by 
the Qur'an (3. 152/145 ff.), although the archers are not specifically 


God made good His promise to you when ye were sweeping them 
away by His permission, until when ye flinched and vied in withdrawing 
from the affair (or, disputed with one another about the affair), and 
disobeyed after He had shown you what ye love; then He turned you 
from them, that He might try you . . . when ye were making for the 
skyline without turning aside for anyone though the Messenger was 
calling to you from behind you. . . . 

If this, then, was the official account, can it be accepted as reliable ? 
So far as the individual incidents go, it is obvious that sources 
1 IH, 575; WW, 119. * WW, i2i f. 


friendly to a man (and these are the majority) will magnify his 
merits and try to hide any faults. For many it is claimed that they 
stood firm with Muhammad and helped him to regain the hill. 
Those who fled to the stronghold of the clan of Harithah are said 
to have turned back to fight, but one may wonder whether they 
were not simply killed as they fled. Usually the discreditable acts 
are anonymous; only the archers who stood firm are named. An 
exception to the rule of anonymity is the suggestion that it was 
Ju'al b. Suraqah who first cried out that Muhammad had been 
killed; but two of the other archers witnessed to his brave bearing, 
and the generally accepted view came to be that the cry had been 
raised by the Devil in Ju f al's form ! Some incidents are remembered 
chiefly for their legal interest, like the case of the man who had just 
become a Muslim and achieved martyrdom without once having 
performed the Worship. 1 The story of Quzman, a non-Muslim 
Medinan who fought bravely on the Muslim side but eventually 
hastened his own death and went to Hell, is frequently referred to 
from the legal and theological standpoint. 2 The ascription of the 
attack on the Muslim rear to Khalid is perhaps due to the hostility 
of the sources to him, as the cavalry is elsewhere said to have been 
under Safwan b. Umayyah. 3 The presence of such tendencies, 
however, in the accounts of the separate incidents does not greatly 
affect the general outline, and we must consider whether this out- 
line is similarly open to criticism. 

The first question to ask is whether the initial Muslim success 
was as extensive as some of the source material suggests, for the 
Muslims are said to have reached the Meccan camp and started 
plundering. This is distinctly doubtful, for the same source material 
also admits that they did not in fact secure any booty. 4 It seems 
probable, then, that, while the Meccan infantry w^s thrown back, 
perhaps in disorder, the Muslim advance was not nearly so exten- 
sive as is claimed. On the other hand, the Meccan retreat rather 
presupposes that the Muslims had been definitely superior when 
fighting on foot under equal conditions. 

The next point is whether the plundering instincts of the archers 
were the main reason that the Muslim rear was exposed to the 
Meccan cavalry. The words 'what ye love' in the Qur'anic passage 
could certainly refer to booty. It seems not impossible, however, 

1 WW, 124. 2 Ibid. 109; cf. al-Bukhari, Qadar, 5 (iv. 253); &c. 

3 WW, 108; WK, 219. 4 WW, in; contrast 112. 


that they were simply overrun by a well-led charge while still in 
position (as the sources admit happened to the remnant), but their 
own slackness when they saw the good fortune of their side may 
have contributed. It is curious that the man to whom the rumour 
of Muhammad's death is attributed should have been one of 
'Abdallah b. Jubayr's band of archers. Moreover, it is conceivable 
that this flank attack was not the sudden inspiration of a moment 
but part of the Meccan plan of battle. The Meccan foot were too 
undisciplined to carry out a planned withdrawal in the face of 
enemy pressure, but those who directed the battle may well have 
foreseen the likelihood of such a retreat and held the cavalry 
in readiness for just such an attack as they made. The fact 
that the Meccans had dug trenches somewhere might indicate a 
deliberate attempt ^o cause confusion in the Muslim ranks, but 
the references are too slight to permit a definite opinion. 1 In the 
light of this total picture it is impossible not to entertain the sus- 
picion that the Meccan success was largely due to skilful general- 
ship. Khalid b. al-Walid is one of the great generals of all time, and 
presumably made his contribution at the council of war before the 
battle, but the supreme commander, Abu Sufyan, and the other 
leaders, were by no means devoid of military skill. 

We may admit, then, that the official Muslim account has to be 
modified along these lines. From the standpoint of Muhammad, 
however, there is a sens$ in which it is true. There had been a great 
upsurge of confidence in their military superiority among the 
believers after Badr. Qur'an 8. 65/66 may be taken as an illustration 
of their general attitude : 

O thou prophet, stir up the believers to fight : 'If there be twenty of you 
who endure, they will overcome two hundred, and if there be a hundred 
of you they will overcome a thousand of those who have disbelieved ' 

Muhammad doubtless realized that there was an element of 
exaggeration in this, as is shown by his hesitation about going out to 
meet the Meccans, but probably felt that it was necessary to speak 
thus in order to counter the long-standing reputation of Quraysh 
and strengthen the morale of the Muslims. This may be the impli- 
. cation of Qur'an 3. 126/122 (though the primary reference is to the 
promise of angels) : 

God only set that forth as good news for you, that your hearts thereby 
might be at peace. 

1 Ibid. 117, &c. 


It is interesting that 8. 66/67, which must have been revealed some 
time after verse 65/66, reduces from a thousand to two hundred the 
number of pagans a hundred Muslims can overcome. Now after 
Uhud Muhammad had to explain how it was that the Muslims 
had not overcome Quraysh. He could not consistently say that it 
was due to the superior numbers of the enemy ; and indeed that 
was not true since to begin with the Muslims had shown them- 
selves capable of dealing with (presumably) larger numbers. It 
would have lowered Muslim morale to attribute it to the fact that 
the Meccans had a large force of cavalry while the Muslims had 
none. Such indications as are given in al-Waqidi's casualty list 
suggest that most of the Muslims who fell were struck down by 
horsemen. But Muhammad had been aware of the danger from 
the cavalry and had made allowance for it in his plans. As the 
matter appeared to Muhammad, then/ the unfortunate result of 
the battle was due neither to the Meccans* superior numbers nor 
to their cavalry (nor to their generalship, he would doubtless have 
added), but to the combination of indiscipline and love of plunder 
among the Muslims. As the Qur'an put it, 'ye flinched . . . and 
disobeyed after He had shown you what ye love*. And this decline 
in the military qualities of the Muslims is no doubt connected with 
their increase in numbers. During the past year Muhammad had 
been aware that he would need a far larger force than that at Badr, 
and he had not turned away men attracted to Islam by prospects 
of booty. Because of this there was serious indiscipline at Uhud 
and lack of steadfastness not merely among the archers but in 
various other ways. This was therefore the point on which he had 
to concentrate as he prepared for the next round of the struggle. 
His rejection of the Jewish contingent which offered to fight for 
him is, if. authentic, perhaps partly due to the. awareness that 
morale was not as good as it ought to be, and that the Jews might 
set a bad example. 

Finally, the common assumption that Uhud was a great defeat 
for the Muslims and a great victory for the Meccans must be con- 
sidered. For the Muslims Uhud was certainly a setback. They had 
over seventy killed, including some old and trusted followers of 
Muhammad, and his father's brother, Hamzah; and among the 
Arabs the loss of each individual tended to be felt. More serious, 
however, was the ideological aspect. Lower morale among the 
combatants was dealt with by the official account of the battle. 


Muhammad's earlier claim that Badr was a sign of God's favour 
raised theological difficulties, which his opponents in Medina and 
elsewhere were not slow to press. Did Uhud not show that God 
favoured the Meccans, and that Muhammad was no prophet? 1 
The theological problems are referred to in various passages of 
the Qur'an : 

(Recall) when ye were making for the skyline without turning aside 
for anyone though the Messenger was calling to you from behind 
you; so He recompensed you with distress upon distress; in order that 
ye may not grieve for what ye have missed or for what has befallen 
you (3. IS3/H7-) 

What befell you on the day when the two hosts met was by the per- 
mission of God, and in order that He might know the believers and in 
order that he might know the hypocrites. . . . (166/160.) 

Let not those wKo Have disbelieved think that such respite as We give 
them (sc. tht Meccans) is for their good; We give them respite simply 
that they may increase in guilt, and for them is a punishment humiliat- 
ing. God is not one to leave the believers in the situation in which ye are 
until He distinguishes the bad from the good (sc. till doomsday). 
(173/172 f.) 

Western scholars have sometimes thought that the sources try to 
hide the full extent of the disaster at Uhud. Scrutiny suggests, 
however, that the opposite is rather the case, and that the Muslims 
themselves paint Uhud^in gloomier colours than it merits. This 
might in part be a reflection of the animosity of the Ansar (who 
were almost the sole sufferers, losing seventy men to the Emigrants' 
four) against Quraysh and especially the ancestors of the Umayyad 
dynasty. The passages from the Qur'an, however, give grounds 
for another explanation in that they show that after Uhud the 
average Muslim, who was more sharply aware of the loss of life 
than of the wider strategic context, was plunged into profound 
spiritual chaos, since some of his cherished beliefs had been 
shattered. His gloom was not lessened by certain events of the 
next few months, when nomads, doubtless responsive to Meccan 
propaganda, showed their contempt for Muhammad by deliber- 
ately shedding Muslim blood. 

If Uhud was not an out-and-out defeat for the Muslims, still 
less was it a Meccan victory. The Meccan strategic aim was the 
destruction of the Muslim community and nothing less, and they 

1 WW, 145- 


had fallen far short of this. For many of the Meccans the conscious 
motive was revenge for the blood shed at Badr; and, if we take 
the lower figure of about fifty for Meccan dead at Badr, then the 
Muslims killed at Badr and Uhud together are slightly more than 
the Meccans killed in the two battles (though with the higher 
figure of seventy Meccans killed at Badr the total Muslim dead 
are slightly fewer). Quraysh as a whole then had had its revenge, 
even if some individuals were not yet satisfied, as subsequent 
events showed. But at best Quraysh had merely taken a life for 
a life, whereas they had boasted they would make the Muslims 
pay several times over. In more general terms the implication of 
Uhud was that Muhammad was almost able to hold his own 
against Quraysh, and that Quraysh were not capable of doing much 
more than holding their own against Muhanynad. What humilia- 
tion for the proud merchant princes oT Mecca who recently had 
thought that they had all western Arabia under their control ! 

Why, then, since this was so, did Abu Sufyan withdraw from 
Medina without pressing home his advantage? He at least must 
have been aware of the strategic necessity of destroying Muham- 
mad's power, and he apparently knew before he left the battlefield 
that the claim of Ibn Qamiyah to have killed Muhammad was 
false. 'Amr b. al-'As (as reported by al-Waqidi) sums up the 
position from the Meccan standpoint : 

When we renewed the attack against them, *ve smote a certain number 
of them, and they scattered in every direction, but later a party of them 
rallied. Quraysh then took counsel together and said, The victory is ours, 
let us depart. For we had heard that Ibn Ubayy had retired with a third 
of the force, and some of the Aws and the Khazraj had stayed away 
from the battle, and we were not sure that they would not attack us. 
Moreover we had a number of wounded, and all our horses had been 
wounded by the arrows. So they set off. We had* not reached ar- 
Rawha' until a number of them came against us, and we continued on 
our way. 1 

In other words, the Meccans were not in a position to do anything 
further. They could not with any hope of success attack either 
Muhammad or the strongholds constituting the main settlement 
of Medina. It is conceivable that they hoped to win over Ibn 
Ubayy by diplomacy, and so did not want to attack him; but this 
is not likely. At the same time they had been roughly handled by 

1 WK, 29i;cf.WW, 138. 

ii. 3 THE BATTLE OF UtfUD 29 

the Muslims; their infantry had been proved inferior, and the 
horses to which they owed their success were temporarily out of 
action. Their morale must now have been low, and their with- 
drawal tends to confirm the Muslim claim to have been victorious 
at first. To return home was clearly the wisest course. 

Muhammad and the Muslims returned to their homes late on 
the day of the battle, after burying the dead. Overnight Muham- 
mad had time to reflect on the position and realized, if he had not 
done so already, that he had suffered no irretrievable disaster and 
that much depended on his actions in the immediate future. On 
the following morning, therefore, he summoned those who had 
been with him at Uhud to set out in pursuit of the retreating 
Meccans. It was the normal and expected thing for an Arab to do 
when he had been the victim of a raid. Muhammad had presum- 
ably no intention of attacking the Meccans, any more than they 
can have thought seriously of attacking him. It was an act of 
defiance and at the same time a sufficient show of strength to deter 
the Meccans from returning to the attack. The latter had appar- 
ently spent the Saturday night at Hamra' al-Asad, a few miles 
from Medina, and thither the Muslims proceeded on the Sunday, 
and camped for three or four days. Contact was not made with the 
enemy, though they were apparently still in the vicinity. To make 
his demonstration more impressive Muhammad had his men work 
hard collecting wood bjr day and lighting fires by night. A friendly 
nomad of Khuza'ah helped to lower the morale of Quraysh by 
exaggerating the number of the Muslims. Quraysh seem to have 
remained near by for a day or two they must not appear to be 
running away from a renewal of the fighting and Abu Sufyan 
tried to spread disquieting rumours among the Muslims. Muham- 
mad, however, (lid not flinch. The enemy made no attempt to bring 
him to battle, and eventually continued on their way to Mecca. 


(a) The Meccans' last chance 

Long before they reached Mecca Abu Sufyan and the other leaders 
must have realized that their position was critical. They had made 
a great effort and had not succeeded. Unless they could do some- 
thing much better they were faced with disaster. For the expedition 
of Uhud they had collected all the available men from Quraysh 


and the surrounding tribes friendly to them. The only possibility 
of raising a more powerful army was to attract the active sup- 
port of some of the great nomadic tribes to the east and north- 
east of Medina, using propaganda about Muhammad's weakness, 
memories of the prestige of Quraysh, promises of booty, and even 
straight bribes. To this task Quraysh now devoted their energies, 
as we may surmise from the incidents that have been recorded and 
from the account of the army they brought together in the year 

627/5- 1 

It was not only among tribes like Sulaym and Ghatafan that 
Quraysh tried to make an impression. The chief of B. Damrah 
(a little north-west of Mecca) had been led to believe that Muham- 
mad was 'finished* after Uhud, and was greatly surprised to see 
the strong Muslim force at Badr in April 6^6 (xi/4). 2 However, 
the tribes round Medina already friendly to Muhammad seem to 
have stood firm, and numbers of individuals from more distant 
tribes were now attaching themselves to him. 

By their exaggerations the Meccans probably did themselves 
more harm than good in the end, for the appearance of a strong force 
of Muslims soon showed up the lies. Muhammad's policy in the 
two years following Uhud was, as far as possible, to forestall hostile 
moves against Medina. As soon as he heard of a concentration of 
tribesmen threatening Medina and he had a good information 
service he sent out an expedition to brjak it up. Such was the 
raid to Qatan against B. Asad led by Abu Salamah with 150 men 
in June 625 (i/4). Exactly a year later Muhammad himself led 
a similar raid to Dhat ar-Riqa' against B. Anmar and Tha'labah 
with 400 (or 800) men. In these raids little seemed to be accom- 
plished apart from the capture of a small quantity of booty; but 
their effect as demonstrations of Muslim strength was important, 
and they doubtless made it much more difficult for the Meccans 
to organize their great confederation. 

Another measure employed by Muhammad where opportunity 
was given to him was assassination. In the period between Uhud 
and the siege of Medina there are two instances of this (according 
to the more probable dating). Sufyan b. Khalid b. Nubayh, chief 
of B. Lihyan, a branch of Hudhayl, was killed by 'Abdallah b. 
Unays (a confederate of B. Salimah of the Khazraj belonging to 

1 IH, 638-68; WW, 149-90; Tab. 1431-65; contrast Caetani, i. 568-613. 

2 WW, 169. 


B. Juhaynah). 1 Again, after the expulsion of B. an-Nadir from 
Medina and their settlement in Khaybar, one of their leaders, 
Abu Rafi' Sallam b. Abi '1-Huqayq, who was engaged in anti- 
Muslim intrigues with B. Ghatafan and other tribes in the vicinity, 
was assassinated by a group of five Muslims from Medina, including 
the same 'Abdallah b. Unays. 2 To such envoys Muhammad gave 
permission to say what they liked about himself and to pretend 
that they wanted to fight against him. Thus any would-be enemy 
had to be constantly on his guard against tricks of this sort, just as 
Muhammad also had to be constantly on guard against the enemy's 
wiles and ruses. Two happenings which incidentally illustrate this 
point are worthy of separate consideration; they are rather different 
from the other 'expeditions' of the period after Uhud. 

(b) Bi'r Maiinate 

The disaster at the well of Ma'unah has been given overmuch 
prominence by some Western biographers of Muhammad. There 
was certainly considerable loss of life for the Muslims, and some 
encouragement for other enemies to take similar measures. But it 
did not raise theological difficulties as the battle of Uhud had 
done ; and in the total strategic picture of the period after Uhud 
it is hardly noticeable, except as an instance of the difficulties with 
which Muhammad had constantly to contend. 

The story is that a leading man of the tribe of 'Amir b. Sa'sa'ah, 
Abu '1-Bara' 'Amir b. MSlik, was invited by Muhammad to become 
a Muslim and showed some readiness to do so provided there was 
sufficient support in his tribe. To obtain this he asked Muhammad 
to send missionaries to the region where its pasture-grounds were, 
and undertook to act as their protector (jdr). Muhammad was 
dubious about the proposition but sent 40 (or 70) young men (but 
even 40 may be'an exaggeration as the names of less than 20 are re- 
corded). Another leader of B. 'Amir, 'Amir b. Tufayl, was appar- 
ently hostile to Muhammad, had the envoy with Muhammad's 
letter killed, and tried to get his tribe to attack the Muslim party; 
but the tribe insisted on observing the protection given by Abu 
'1-Bara'. 4 'Amir b. Tufayl, however, persuaded some neighbouring 
clans of B. Sulaym to attack, and the Muslims were killed, apart 
from two. One of these was seriously wounded and left on the 

1 IH, 981-3 (cf. 310, 501) ; WW, 224 f. ; Usd, iii. 1 19 f. For the date cf. Caetani, 
i. 578-80. 2 IH, 981; WW, 170-2. 3 IH, 648-52; WW, 153-6. 

4 Cf. p. 97 below. 


field for dead. The other, 'Amr b. Umayyah al-Kinam, was taken 
prisoner, but set free because of some relationship to his captors. 
On the way home he met two members of B. 'Amir and killed them. 
When 'Amir b. Tufayl asked Muhammad for blood-money for 
these two members of his tribe, he was paid it, as the tribe was in 
alliance with the Muslims. 

The puzzle in this account is that Muhammad pays blood- 
money to 'Amir b. Tufayl for two men and does not make any 
counter-claim for nearly forty Muslims for whose death 'Amir b. 
Tufayl had just been responsible. It has been suggested that the 
Muslims had done something which caused them to forfeit the 
right to blood-money; but this is improbable. An explanation that 
does little violence to the existing texts is that, while 'Amir b. 
Tufayl had encouraged the Sulami clans to massacre the Muslims 
and so was morally responsible, he Was not tfieir leader in any 
sense and so not technically responsible; it was probably also they 
who killed the letter-carrier. 

This explanation of the blood-money can be made part of a con- 
sistent account of the whole affair. Abu '1-Bara', who was the uncle 
of 'Amir b. Tufayl, had a strong following within the tribe which 
disapproved of 'Amir. His appeal to Muhammad was at bottom 
an appeal for help against rivals within the tribe. Muhammad, 
anxious to bring B. 'Amir to his own side, decided to interfere in 
the internal politics of the tribe, though Jie realized the riskiness 
of doing so. When 'Amir b. Tufayl discovered he was not strong 
enough to bring the tribe to disown Abu 'l-Bara"s 'protection', 
he persuaded his neighbours of B. Sulaym to attack the Muslims, 
and doubtless gave them help by way of information. 

It was, of course, shameful for B. 'Amir to allow men under its 
protection to be killed. The Muslim poet, Ka'b. fy. Malik, is quite 
clear on the point: 'You left your proteges to Banu Sulaym, in 
abject weakness fearing their attack. . . .'* On the other hand, 
Hassan b. Thabit, who was friendly with the descendants of Abu 
'1-Bara', blamed rather the treachery of 'Amir b. Tufayl: 'Death 
came upon him (one of the victims) through league with a tribe 
whose league-making was rendered ineffectual by treachery. . . .' 2 
Muhammad could not abandon his alliance with B. 'Amir without 
giving up many hopes, but nothing prevented him praying that 
God would punish 'Amir. 3 Perhaps Muhammad's attempt to make 
1 IH, 652. 3. 2 Ibid. 651. 18. 3 WW, 155. 


the Nadir, who were confederates of B. ' Amir, contribute a portion 
of the blood-money of the two f Amiris was in part a way of paying 
back 'Amir b. Tufayl. 

(c) Ar-Raj? 1 

The small expedition of seven men which met with disaster at 
ar-Rajr (on the Najd road) raises several difficulties, but our 
information permits sufficient accuracy to make the event a useful 
illustration of the conditions in which Muhammad and the 
Muslims lived. 

One version, that of 'Urwah, was that Muhammad sent out the 
seven men to get information about Quraysh presumably about 
their intrigues with the great nomadic tribes. The common version, 
however, is that B. Lihyan wanted to avenge the assassination of 
their chief at Muhammad's instigation, and bribed two clans of 
the tribe of Khuzaymah to say they wanted to become Muslims 
and ask Muhammad to send instructors. At ar-Raji f the would-be 
converts left the 7 Muslims alone and informed their enemies, who 
surrounded them, killed 4 who resisted and took the other 3 
prisoners, after making fair promises to them. One of the 3 escaped 
from his bonds on the way to Mecca and died sword in hand, but 
the other 2 were sold in Mecca to relatives of men killed at Badr; 
after the sacred month was over they were taken outside the sacred 
area, summoned to recant from Islam and when they refused put 
to death not altogether painlessly. 

The difficulty lies in the motive of B. Lihyan, for a commonly 
expressed view is that the assassination of the chief did not take 
place till the year 627/5 or even later. Though some sources give the 
date as the 35th month after the Hijrah (June 625 = i/4), others 
say it was the 54th or 55th month (about January 627 = viii/5). 
In view of the alleged motive for the attack on the Muslims at 
ar-Raji c in July 625 (ii/4), however, and in view of the fact that 
Muhammad cursed B. Lihyan along with those responsible for 
Bi'r Ma'unah, 2 it is probable that the assassination took place in 
the 35th month before the affair of ar-Rajf . The two explanations, 
moreover, may well be complementary. The seven may have been 
both spies for Muhammad and instructors for Arab tribes. The 
incident is trivial, but it shows how the Arabs expected men to 

1 IH, 638-48; WW, 156-60; Caetani, i. 581-4. 

2 WW, i 5 5;WK, 341. 



take care of themselves and to be on their guard against dupli- 
city. The treatment of prisoners suggests that feelings were now 
pretty high on both sides. 

(d) Growing Muslim strength 

The battle of Uhud, it has been maintained, was not a great 
military disaster for the Muslims, though it caused them some 
theological heart-searchings. The expeditions to Hamra* al-Asad 
and Qatan (March and June 625) were not unsuccessful. The 
misfortunes at Bi'r Ma'unah and ar-Raji e (about July 625) may have 
caused temporary gloom in Medina, but for the rest of the period up 
to the siege the tide seemed to be turning in favour of the Muslims. 

The expulsion of the tribe of an-Nadir from Medina in August 
625 (iii/4) will be considered more fully in. connexion with the 
internal politics of Medina, but it has also some connexion with 
Bi'r Ma'unah, and may further have been intended to cheer up the 
Muslims after the news of the two disasters had reached them. 
Certainly a few months later (April 626 = xi/4) Muhammad was 
able to raise a force of 1,500 men and 10 horses to go to Badr. 1 
This was by far the largest number of men he had so far collected. 
The story (which may or may not be true) is that before Abu 
Sufyan left the field at Uhud he shouted to the Muslims, 'We 
would like to meet you at Badr next year! 1 , and 'Umar replied in 
Muhammad's name, 'We will be there!' Each side tried to scare 
away the other by exaggerated accounts of its strength. Eventually 
the 1,500 Muslims spent the eight days of the market and fair at 
Badr, and did not come into contact with the 2,000 men and 50 
horses which Abu Sufyan brought out to meet them. Presumably 
both sides wanted to avoid fighting and merely to make a demon- 
stration of strength. The Muslims, however, seem to have had 
the better of their opponents in this matter, and the tribes of the 
coastal region were suitably impressed. 2 It is possible that the 
Muslims circulated the story of the rendezvous merely to discredit 
their opponents. 

During the next few months Muhammad acted with severity 
whenever he heard of men massing with hostile intentions against 
Medina. As already mentioned, Abu Rafi' was assassinated (May 
626). The tribes of Anmar and Tha'labah were raided at Dhat 
ar-Riqa c (June 626). Another large force of 1,000 men broke up 

1 IH, 666-8; WW, 167-70. 2 WW, 169. 


what was possibly an enemy concentration at Dumat al-Jandal 
(August and September 626). This expedition receives scant notice 
in the sources, but in some ways it is the most significant so far. 
As Dumah was some 500 miles from Medina there can have been 
no immediate threat to Muhammad, but it may be, as Caetani 
suggests, 1 that communications with Syria were being interrupted 
and supplies to Medina stopped. It is tempting to suppose that 
Muhammad was already envisaging something of the expansion 
which took place after his death. It may be, however, that his 
primary aim was to deter these northern tribes from joining the 
Meccan grand alliance against him; but what he now learnt about 
conditions in the north may have shown him the possibility of 
expansion in this direction. Certainly his rapid march with such 
a large force must h^ve impressed all who heard of it. Despite Uhud 
Muslim strength was clearly increasing. 

The expedition to al~Muraysi' 2 against B. al-Mustaliq, a branch 
of Khuza'ah, is placed by al-Waqidl in viii/5 (== January 627), 
though Ibn Ishaq places it later after the siege of Medina. The 
chief of the tribe or clan was said to be arming the men for an 
attack on Medina (perhaps in concert with the expedition the 
Meccans were planning). Muhammad attacked the small group 
unexpectedly with overwhelming force, and after only a brief 
resistance all were taken prisoner. The presence of Hypocrites in 
the Muslim force and the quarrels to which this led belong rather 
to the internal history of Medina. As al-Muraysi* is near the Red 
Sea coast, north-west of Mecca, the expedition is an indication of 
how Muhammad's sway was encroaching on the sphere where 
Mecca recently had been supreme. 

Thus in the period between Uhud and the siege of Medina, 
while Muhammad was unable to prevent the Meccans forming 
a confederation against him, he probably stopped many from joining 
it, and he certainly increased the forces at his own disposal. From 
the purely human point of view he could not regard the threatened 
attack without anxiety, yet he had also good grounds for hope. 


The siege of Medina, 3 known to Muslims as the expedition of 
the Khandaq or Trench, began on 31 March 627 (8/xi/5) and 

1 i. 597. 2 IH, 725-40; WW, 175-90; Caetani, i. 599-606. 

s IH, 668-713; WW, 190-210; Tab. 1463-85; Q. 33. 9-27; Hamidullah, 
Battlefields of Muhammad, 25-30. 


lasted about a fortnight. 1 It was the supreme effort of the Meccans 
to break Muhammad's power. For it they had gathered a vast con- 
federacy, including some of the nomadic tribes in no way subject 
to them. The Jews of an-Nadir, now in exile at Khaybar and eager 
to regain their lands at Medina, had much to do with the collecting 
of the confederacy; half the date harvest of Khaybar was promised 
to B. Ghatafan if they would join in the attack. 2 

The Meccan confederacy had 10,000 men in three (or perhaps 
two) armies. Quraysh and their closer allies constituted one army 
of 4,000 men; Ghatafan was the leading tribe in another, and 
Sulaym in a third; but the latter seem toiiave shared a camp site 
with the Ghatafan group. 3 Three branches of Ghatafan, Fazarah, 
Ashja', and Murrah, supplied respectively 1,000, 400, and 400 men, 
and Sulaym 700. That makes only 6,500*, and the other tribe 
mentioned as participating, Asad, could not account for 3,500, but 
it is idle to speculate about the discrepancy. Quraysh themselves 
had 300 horses, and there was a like number in the army of 

To oppose this enormous force Muhammad could count on 
about 3,000 men, that is, practically all the inhabitants of Medina 
with the exception of the Jewish tribe of Qurayzah, who seem to 
have tried to remain neutral. There were some Medinans in league 
with the Meccans, but they were presumably, like Abu 'Amir 
ar-Rahib, exiled from Medina for the- time being. There was a 
considerable degree of unity among the Muslims, but some of the 
Hypocrites were critical of Muhammad's methods and sceptical 
of a successful result. The Qur'an makes it clear that a number of 
them would have been only too glad to be out of the fighting and 
might even have gone over to the enemy had there been a favour- 
able opportunity. 4 

As in the campaign of Uhud the enemy approached Medina by 
Wadi' l- c Aqiq, and camped partly there and partly beside mount 
Uhud. The latter arrangement may have been intended to prevent 
the Muslims obtaining the strong position there which had been 
so advantageous to them on the previous occasion. Such a pre- 
caution, however, was vain, for Muhammad had adopted another 
form of defence, indeed, one hitherto unknown in Arabia. Wherever 
Medina lay open to cavalry attack he had dug a trench, the 

1 WW, 190. * Ibid. 191. 

3 Ibid. 191 n. 4 33. 9-25, esp. 12-15. 


Khandaq. The idea may have come from Persia, and the Persian 
convert Salman is credited with an important share in the detailed 
planning. The work was set afoot as soon as it was known that the 
Meccans had started out it would have been difficult to rouse 
enthusiasm for it earlier and most of the Muslims worked hard 
at it for six days until it was completed. Muhammad established 
his headquarters on mount Sal', and from this spot could presum- 
ably command a view of the whole northern front. Had the enemy 
crossed the trench, Sal' would presumably have given the de- 
fenders some of the advantages they had at Uhud. 

On one occasion a small party of horsemen crossed, but they 
were too few to accomplish anything of importance, and in the 
end retired with the loss of two of their number. Apart from this 
the Meccans failed to cross the trench at all. They made several 
assaults by night, but the trench was guarded constantly. To effect 
a crossing the infantry also would have had to engage the Muslims 
at close quarters, and that they seem to have been unwilling to do, 
for they probably regarded the Muslims as more than a match for 
them in hand-to-hand fighting. The one hope of the Meccans 
under these circumstances would have been to make several 
attacks at once. They seem to have hoped to persuade B. Qurayzah 
to attack the Muslims from the south, but nothing came of the 
negotiations. The Muslims were sufficiently well organized and 
sufficiently numerous to contain all assaults from the north. After 
a fortnight spent in this way the Meccans gave up hope of success, 
and the great confederacy split up into its separate contingents, 
and retired. Exceptionally cold weather and a storm of wind gave 
the coup de grace to the morale of the besiegers. Six of the Ansar 
are reported to have been killed and three of the Meccans. 

On the military side the reason for the Meccan failure was the 
superior strategy of Muhammad, and probably also his superior 
information service and secret agents. In particular his adoption 
of the trench was well suited to the circumstances. Meccan hopes 
of victory rested mainly on the superiority of their cavalry, for 
previous battles had shown that the Muslims were likely to over- 
come their opponents in an infantry melee unless heavily out- 
numbered. The trench effectively countered the menace from the 
cavalry and forced the Meccans to fight in conditions where they 
derived little advantage from their 600 horses. 

The Meccans suffered from a further disadvantage. In the 


campaign of Uhud they had arrived at Medina about ten days 
earlier, before the grain was harvested; and the fields had provided 
fodder for the Meccan horses, while the sight of their devastation 
had provoked the Medinans to march out in defence despite the 
disadvantageous circumstances. In 627/5 the g ra i n had been har- 
vested a month before the Meccans arrived probably earlier than 
usual because of Muhammad's foresight and they had great 
difficulty in obtaining fodder for their horses. Moreover, as there 
was no provocation, the Ansar were content to remain behind the 
trench. This point seems to indicate slackness or lack of foresight 
on the part of the Meccan leaders. 

Apart from these purely military considerations, the result was 
due to the relative unity of the Muslims and their discipline, in 
contrast to the lack of cohesion in the confederacy and the lack 
of confidence of the various groups in one another. Of this disunity 
Muhammad's diplomacy took full advantage. The main group of 
nomads in the confederacy, Ghatafan, had only been persuaded 
to come by a bribe, and Muhammad made tentative offers to them 
of a presumably higher bribe if they would withdraw. The report 
is that he offered them a third of the date-harvest of Medina, but 
that at first they demanded a half, and only after some time agreed 
to accept a third ; when this came to the ears of some of the leading 
Medinans they protested that Medina had never before sunk to 
this depth of ignominy, and insisted that the negotiations should 
be broken off. Whatever the exact details may have been, Ghatafan 
had compromised themselves by discussing such matters with 
Muhammad. The whole was a battle of wits in which the Muslims 
had the best of it; without cost to themselves they weakened the 
enemy and increased the dissension. 

The same is true of the intrigues in which the tribe of Qurayzah 
was involved. They seem to have had a treaty with Muhammad, 
though it is not clear whether, in the event of an attack on Medina, 
they were to help him or merely to remain neutral. They are said 
to have supplied the Muslims with some implements for the 
digging of the trench. Later, however, Huyayy b. Akhtab of an- 
Nadir persuaded them that Muhammad was certain to be over- 
whelmed and they changed their attitude. As they would be 
exposed to Muslim retaliation should the confederacy retire without 
destroying Muhammad, they demanded hostages from Quraysh 
and Ghatafan. Negotiations over this were protracted. A secret 


agent of Muhammad's, acting in accordance with hints from him, 
so increased the suspicion with which the different parties viewed 
one another that the negotiations came to nothing, and the 
threatened 'second front* was never opened. The importance of 
this diplomatic success can hardly be overestimated, for an attack 
from the south on the Muslim rear by Qurayzah might have put 
an end to Muhammad's career. 

The break-up of the confederacy marked the utter failure of the 
Meccans to deal with Muhammad. The outlook for them now 
was dismal. They had exerted their utmost strength to dislodge him 
from Medina, but he remained there, more influential than ever 
as a result of the fiasco of the confederacy. Their trade with Syria 
was gone, and much of their prestige lost. Even if Muhammad did 
not attack them, they had no hope of retaining their wealth and 
position; but he might very well use armed force against them, and 
try to annihilate them as they had tried to annihilate him. It would 
be strange if some of the Meccans a practical people had not 
begun to wonder whether it would not be best to accept Muham- 
mad and his religion. 




THE great interest of the period from the end of the siege of 
Medina to the conclusion of the treaty of al-I^udaybiyah is 
that in it new trends become manifest in Muhammad's 
policy. To speak of a reorientation would be to exaggerate, or 
rather to confess that one had failed to understand Muhammad's 
policy so far. Hitherto he had had to concentrate on the struggle 
with Mecca, and it would be natural to suppose that he had had 
no thought beyond the defeat of the Meccans and the conquest 
of their city. Soon after the siege, however, it is ciear that Muham- 
mad's aims are much vaster and more statesmanlike; and when 
one scrutinizes the early history there appear slight indications that 
these wider aims were present all along, or at least since the 
victory at Badr had shown that great changes were possible. The 
study of this period will therefore most profitably be directed to 
attempting to understand the underlying aims of Muhammad's 
overt actions. 

In such a study we necessarily use an analytic and discursive 
mode of thought. Muhammad himself, however, almost certainly 
thought intuitively and not analytically. He was aware of all the 
factors we laboriously enumerate, but, without isolating these in 
his thinking, he was presumably able to decide on a course of 
action that was an adequate response to them. In particular the 
religious aspect of events was the dominant one for Muhammad, 
even where he was most fully aware of political implications ; and 
he would almost certainly have described his supreme aim at this 
period as the summoning of all the Arabs to Islam. The implied 
corollary, namely, the political unity of all the Arabs cannot have 
escaped Muhammad, but it remained in the background. 

To speak at this stage of all the Arabs may seem to be going 
too far ahead, since Islam had touched only a few tribes in the 
neighbourhood of Mecca and Medina. But Muhammad had suffi- 
cient width of vision to look beyond immediate concerns, and it 
would be natural for him to take as his potential unit the Arabian 
peninsula, or rather the totality of tribal groups with some claim 


to the name Arab. On the other hand, the suggestion of some 
Muslim sources, though not the earliest, that he conceived of 
Islam as a universal religion and summoned the Byzantine and 
Persian emperors and other lesser potentates to accept it, is almost 
certainly false. Islam indeed from its beginnings was potentially 
a universal religion, and it is not fortuitous that with the expansion 
of the Islamic state it became in fact a universal religion. But it is 
barely credible that a wise statesman like Muhammad should have 
made this precise appeal at this precise stage in his career; and 
examination shows that the reports of the embassies to the various 
sovereigns are full of inconsistencies. 

The usual account 1 is that on one day shortly after his return 
to Medina from al-Hudaybiyah Muhammad sent out six messen- 
gers with letters for the Najashi or Negus of Abyssinia, for the 
governor of Bostr'a (Busra) to hand on to the Byzantine emperor, 
for the Persian emperor (perhaps sent by way of the Yemen), for 
the Muqawqis or ruler of Egypt, for al-Harith b. Abi Shamir, 
prince of Ghassan, and for Hawdhah b. 'All, chief of Hamfah. 
The messengers who carried the letters are named and the actual 
texts allegedly reproduced. The critical discussions of European 
scholars have shown that, while the story cannot be taken as it 
stands, there is a kernel of truth in it. According to the story 
Muhammad's envoys were favourably received and given presents, 
apart from the one to the Persian emperor. But this is incredible 
if the message was a summons to become a Muslim and accept 
Muhammad as religious leader; we cannot conceive of a Roman 
emperor or a Negus of Abyssinia responding to such a message. 
But, if we admit that the persons named actually carried some 
message from Muhammad to their respective destinations (though 
probably at different dates) and were well received, it is not im- 
possible that the contents of the letters have been somewhat 
altered in the course of transmission. This may be either because 
the details were not known to the messenger (who is the presump- 
tive source of information), or because later developments made 
the message seem trivial and unworthy of a great prophet. On this 
hypothesis we might suppose that, while Muhammad may have 
made some reference to his religious beliefs, the real point was 

1 IS, i/2, 15-18 ( 1-7); commented on by J. Wellhausen, Skizzen und 
Vorarbeiten, iv, Berlin, 1889, 97-102; Caetani, i. 728-39; Buhl, Muhammed, 
' 294-8. Cf. Excursus D, p. 345 below. 


political. Perhaps he proposed a neutrality pact. Perhaps he was 
merely anxious to prevent the Meccans getting foreign help and 
to counteract the effects of the biased accounts they gave of their 
relations with him. It would have been most inappropriate for 
Muhammad at this period to summon these powerful tulers to 
accept Islam. But after the siege of Medina he was sufficiently 
important to have some rudimentary diplomatic contacts with 
them; and that is presumably the truth of the matter. This is in 
keeping with the view that Muhammad was now interested in 
more than the defeat of Quraysh. 

In the alliances made during this period Muhammad does not 
seem to have insisted that those on whom the privilege of being 
his allies was conferred should be Muslims. The words of Ibn 
Sa'd suggest that the treaty with Ashja' was concluded some time 
before they became Muslims; and Caetani has argued that the 
people of Dumat al-Jandal did not cease to be Christians. 1 Other 
possible instances will be mentioned in the next chapter. 

Of the various events placed by al-Waqidi in the year between 
the siege and the expedition to al-Hudaybiyah, one section might 
be described as an aftermath of the failure of the Meccan con- 
federacy at the siege. 2 There was first of all the punishment of the 
Jewish tribe of Qurayzah for their intrigues with the Meccans; 
this will be considered more fully elsewhere. Then there was 
the raid on Muhammad's private herd of camels by 'Uyaynah b. 
Hisn al-Fazarl, who was doubtless annoyed because Muhammad 
had broken off negotiations with him over the withdrawal of 
Ghatafan. The raid was a small affair. Only 40 enemy horsemen 
were involved, and the booty was only 20 milking camels; 8 
Muslims pursued on horseback, recovered half the camels, and 
killed 4 of the raiders for the loss of i of their own number. 
Muhammad, however, seems to have been afraid of a large-scale 
attack, for he collected 500 (or 700) men before following up the 
8 Muslim horsemen, and Sa'd b. 'Ubadah was left on guard in 
Medina with 300 men under arms. A poem by Hassan b. Thabit 3 
suggests that 'Uyaynah hoped to lay Medina waste and obtain 
much booty; this is doubtless poetical exaggeration, but it is an 
indication of what was in the minds of the Muslims. Later in the 
year Zayd b. Harithah had a mishap at the hands of B. Badr b. 

1 IS, i/2. 48 f. ( 92); Caetani, i. 701 ; cf. p. 364 below. 

3 For detailed references see Excursus B, 3 IH, 724. 


Fazarah, a part of 'Uyaynah's tribe, but subsequently avenged it. 

Several of the small expeditions of the period were directed, in 
part at least, against other members of the Meccan confederacy, 
especially Asad and Tha'labah (a part of Ghataian). It appears, 
however, from various remarks in the sources, that rain, and there- 
fore also fodder, was scarce this year, and in particular that Mu- 
harib, Tha'labah, and Anmar had left their usual pasture-grounds 
owing to lack of rain and had come nearer to Medina. It may 
therefore be that the Muslim raids were not so much a requital for 
what had gone before as a warning not to encroach on the lands 
of Medina. The same may be true of the expedition against B. Bakr 
b. Kilab (a part of 'Amir b. Sa'sa'ah of Hawazin), although these 
were, at least potentially, allies of the Meccans. A group of B. 
'Uraynah (which liyed among the same Kilab) came to Medina 
(perhaps in distress through lack of food) and professed Islam; 
because they were suffering from fever they were allowed to go 
to the pasture-grounds of Muhammad's private herd to enjoy the 
plentiful milk there. But when they recovered their strength, they 
killed one of the herdsmen and made off with fifteen camels ; they 
were quickly captured and cruelly put to death. Similar to these 
incidents was the expedition led by Muhammad himself to punish 
B. Lihyan who had treacherously attacked a small Muslim party 
at ar-Raji' in July 625 (ii/4). 

The Jewish participants in the Meccan confederacy continued 
their intrigues. They offered bribes for military help to Arab 
tribes, including that of Sa'd ( ? Sa'd b. Bakr), and a Muslim 
expedition was sent against Sa'd which captured 500 camels and 
2,000 sheep as a warning that it was dangerous to fraternize with 
the enemies of Islam. The Jews themselves were given a similar 
warning by the assassination of their leader, Usayr b. Razim. 

A second group of events during the period under review aimed 
at developing closer relations with the tribes on the road to Syria. 
Attention has already been called to the great expedition to Dumah 
in the latter half of 626 (iii-iv/5). Several events in the year after 
the siege point to an expansion of interest in this direction, but 
our information is tantalizingly meagre. 

A series of incidents involves Dihyah b. Khalifah al-Kalbi and 
B. Judham. Dihyah was sent as envoy to Caesar, that is, presum- 
ably, to the nearest Byzantine governor. (This is probably the 
embassy which is commonly placed among the six.) On his way 


back Dihyah was robbed of the Byzantine presents, which he was 
carrying to Muhammad, by a few members of the tribe of Judham. 
He reported the theft to Muhammad, who sent out Zayd b. 
Harithah to punish the offenders. Meantime, however, other 
members of Judham were persuading the tribe to accept Islam; 
complaints were made to Muhammad that Zayd's punitive mea- 
sures were illegal, and 'All had to be sent to straighten out the 
matter. The details are obscure, but it is clear that at least a section 
of Judham entered into alliance with Muhammad about this time. 

Following upon thik Zayd set out on a trading journey to Syria 
in the course of which he was wounded and robbed by B. Badr 
b. Fazarah. 1 This is the first mention of a trading journey by any 
Muslim from Medina. But the next expedition, one of 700 men 
to Dumat al-Jandal, was led by 'Abd ar-Ratyman b. 'Awf, who 
had the reputation of being the shrewdest merchant and financier 
among the Muslims; at some time or other (probably after Mu- 
hammad's death) he appears as the organizer of a caravan of 500 
camels. 2 On the expedition in 627/6 the prince of Dumat al-Jandal 
made a treaty of alliance with Muhammad, and gave his daughter 
Tumadir in marriage to 'Abd ar-Rahman. 

These scanty details are sufficient to show that Muhammad's 
interest in the tribes on the route to Syria was not simply a matter 
of chance. These tribes were either Christian or had some acquain- 
tance with Christianity, and because of tliis may have been more 
attracted to Islam than the pagans farther south, especially while 
the Persians were occupying Syria. But our sources do not'justify 
the supposition that Muhammad was interested in the northern 
tribes because they showed themselves interested in Islam. It is 
more likely that his interest was due to the importance of the Syrian 
trade in the Meccan economy. By his raids on Meccan caravans 
he had blocked the Meccan path to the north, and alliances with 
the northern tribes would serve to tighten the blockade, if Mu- 
hammad so desired. The expeditions of Zayd and 'Abd ar-Rah- 
man were probably designed to bring part of the Syrian trade to 
Medina. This trade was perhaps more important in the life of 
Medina than our sources indicate. As the population of the oasis 
grew through the attraction of Islam, imports of food would pre- 
sumably be necessary. 

1 IS, ii/i. 65. 16; cf. 6 4 . 16 f.; WW, 236, 238. 

2 IS, iii/i. 93. 17 ff. 


Another point was doubtless present in Muhammad's mind. 
He forbade fighting and raiding between Muslims, and conse- 
quently, if a large number of Arab tribes accepted Islam or even 
merely accepted Muhammad's leadership, he would have to find 
an alternative outlet for their energies. Looking ahead, Muhammad 
probably realized that it would be necessary to direct the predatory 
impulses of the Arabs outwards, towards the settled communities 
adjacent to Arabia, and he was probably conscious to some extent 
of the development of the route to Syria as a preparation for 

Thirdly, in contrast to all this activity to the north and east of 
Medina, there was no attempt to attack Mecca directly. On the 
expedition against Banu Libyan Muhammad seems to have made 
a feint against lyte^ca; 1 but it cannot have caused more than a 
temporary flurry. More serious was the attack by Zayd with 170 
men on the caravan returning to Mecca from Syria by the 'Iraq 
road. The goods, including silver belonging to Safwan b. Umay- 
yah, were all captured, and a number of prisoners made. The 
lenient treatment of one of these, however, may be the expression 
of a new policy of leniency directed to winning over the Meccans. 
This man was Abu 'l-'As b. ar-Rabi', husband of Muhammad's 
daughter Zaynab ; he sought and publicly received her protection 
(jizudr), although this was perhaps contrary to the constitution. 
Muhammad denied prjor knowledge of Zaynab's declaration but 
asserted that he was ready to uphold it, and Abu 'l-'As conse- 
quently received back his property that was among the booty. 

From a consideration of these events of 627/6, together with 
what happened subsequently, it seems clear that Muhammad was 
not preparing for a direct assault on Mecca. His policy was instead 
to weaken Mecca by preventing the movement of caravans to and 
from Syria, while at the same time increasing the number of tribes 
in alliance with himself and consolidating the strength of this 
group. The conversion of Ashja c , one of the weaker tribal bodies 
joining in the siege, showed that the Meccans and their con- 
federates were not able to provide for their members as Muham- 
mad was. The peace of Islam, as administered by the iron hand of 
Muhammad, would bring prosperity for the Arabs, but only if the 
means of subsistence were correspondingly increased. But the 
number of camels and sheep the desert could support could not 

1 IS, ii/i. 57. 16-18; contrast WW, 227 = IS, ii/i. 57. 6. 


be greatly increased. Consequently the Islamic state was under 
the necessity of constantly expanding its sphere of influence. From 
now on Muhammad seems to be more concerned with the positive 
building up of strength and prosperity than with the negative aim 
of defeating the Meccans. Soon it becomes apparent that the 
Meccans have an important part to play in the positive side of 
his policy. 


This series of events was brought to a close, which was indeed 
a consummation, by Muhammad's expedition to al-Hudaybiyah; 
and an endeavour must now be made to elucidate this event. The 
outline of the story as given in the sources 1 is that as the result 
of a dream Muhammad decided to go on pilgrimage ('umrah) to 
Mecca. He called on the Muslims (and perhaps others) to join 
him and to bring animals for sacrifice. Eventually he set out with 
from 1,400 to i, 600 men, among whom were nomads of Khuza'ah, 
though other tribes who might have been expected to take part 
made excuses for abstaining. When the Meccans heard of this 
approaching force, they assumed that Muhammad's intentions 
were hostile, and sent out 200 cavalry to bar the way. By taking 
an unusual route across difficult hilly country Muhammad circum- 
vented the Meccan cavalry and reached al-Hudaybiyah on the 
edge of the sacred territory of Mecca. At this point his camel 
refused to go farther, and he decided it was time to halt. 

The Meccans threatened to fight if Muhammad tried to perform 
the pilgrimage. Messengers came and went between them and the 
Muslims, and eventually a treaty was agreed on. This year the 
Muslims were to retire, but in the following year the Meccans 
were to evacuate their city for three days to enable the Muslims 
to carry out the various rites connected with the pilgrimage. At 
a moment when it looked as if the negotiations would break down, 
the Muslims made a pledge to Muhammad known as the Pledge of 
Good Pleasure, or the Pledge under the Tree. On the conclusion 
of the treaty, Muhammad killed his sacrificial animal and had his 
hair shaved, and the Muslims, apparently after some hesitation, 
followed his example. Then they set off home. 

The dream mentioned by al-Waqidi (though not by Ibn Hisham 
and Ibn Sa'd) may be accepted as fact in the light of the Qur'anic 

1 IH, 740-55; WW, 241-64; IS, ii/i. 69-76; Tab. 1528-51. 


verse, 'Assuredly God hath given to His messenger a true and 
right vision 1 ; 1 but the account of the contents of the dream is 
probably influenced by later events. It was doubtless by a dream 
that the idea first came to Muhammad of making the pilgrimage, 
and he was naturally puzzled when what he regarded as a Divine 
promise was not fulfilled. The idea, however, must also have 
commended itself to him for practical political reasons. He can 
hardly have hoped to conquer Mecca, for he must have known 
that the morale of the Meccans was still good, and his force was 
too small to overcome them in battle. His primary intention was 
no doubt simply what he said, to perform the pilgrimage; but this 
had certain political implications, and it was probably in these 
that he was chiefly interested. The performance of the pilgrimage 
would be a demonstration that Islam was not a foreign religion but 
essentially an Arabian one, and in particular that it had its centre 
and focus in Mecca. A demonstration of such a kind at such a time 
would impress upon the Meccans that Islam was not a threat to 
the religious importance of Mecca. It would also suggest that 
Muhammad was prepared to be friendly on his own terms, of 

Unfortunately, however, Muhammad's proposed action, if 
carried out, would make it appear that Quraysh were too weak to 
stop him. It was, of course, one of the sacred months in which 
there was supposed to b$ no bloodshed, but Muhammad had not 
shown himself specially observant of sacred times, and was clearly 
relying, not solely on the sanctity of the season, but partly on the 
number of his followers. In the light of the failure of Quraysh in 
besieging Medina, the triumphant progress of Muhammad through 
Mecca would look bad. It was therefore understandable that they 
should decide to oppose. The compromise eventually agreed on 
saved their face and their prestige, while Muhammad obtained all 
that he really wanted. 

The terms of the treaty are given as follows : 

In Thy name, O God. This is the treaty which Muhammad b. 'Abdal- 
lah made with Suhayl b. c Amr. They agreed to remove war from the 
people for ten years. During this time the people are to be in security 
and no one is to lay hands on another. Whoever of Quraysh comes to 
Muhammad without permission of his protector (or guardian), Muham- 
mad is to send back to them; whoever of those with Muhammad comes 

1 48. 27. 


to Quraysh is not to be sent back to him. Between us evil is to be 
abstained from, 1 and there is to be no raiding or spoliation. Whoever 
wants to enter into a covenant and alliance with Muhammad is to do so ; 
and whoever wants to enter into a covenant and alliance with Quraysh 
is to do so. ... You are to withdraw from us this year and not enter 
Mecca against us ; and when next year comes we shall go out in front 
of you and you shall enter it (Mecca) with your companions and remain 
in it three days; you shall have the arms of the rider, swords in scab- 
bards ; you shall not enter it bearing anything else. 2 

This is probably not an exact reproduction of the original text 
of the treaty, in view of the abrupt changes of person, 3 but it may 
be accepted as an adequate account of the provisions. It gives 
some satisfaction to both parties. The abandonment of hostilities 
for ten years expresses Muhammad's peaceful intentions towards 
Mecca, and gives Quraysh a respite from tne "desperate struggle 
against his growing power. The postponement of the pilgrimage 
rites by the Muslims saved the face of Quraysh, but Muhammad 
achieved his aim of demonstrating his intentions and attitudes by 
the permission to perform pilgrimage in the following year ; indeed 
he had largely achieved it by the very conclusion of the treaty. 
The clause about returning persons under protection (chiefly 
minors and clients presumably) was a concession to the feelings of 
Quraysh which cost the Muslims little. The son of one of the 
Meccan negotiators is said to have come to Muhammad while his 
father was still with him and to have been told he must remain in 
Mecca; but Muhammad made the other two negotiators present 
guarantee his safety in view of his strained relations with his 
father. The fact that this clause was not reciprocal is perhaps 
mainly an expression of Muhammad's belief in the superior 
attractiveness of Islam. 

The remaining clause was apparently one on which Muhammad 
set considerable store, for on his way to Mecca he is said to have 
told one of his messengers that he was ready to make peace with 
Quraysh if they would allow him a free hand with the nomadic 
tribes. The clause suggests that the two sides are being treated 
equally, and in a sense it is a recognition by Quraysh of Muham- 
mad's equality with themselves. In fact it is a concession by Quraysh, 

1 The meaning of this phrase is obscure; cf. E. W. Lane, Arabic-English 
Lexicon, London, 1863-93, s.v. 'aybah. 
* IH, 747 f.; cf. IS, ii/i. 70 f.; WW, 257. 
3 Cf. Caetani, i. 718. n. 2. 


permitting tribes to abandon the Meccan alliance for that of 
Muhammad ; and Khuza'ah speedily made the exchange. 

This is part of Muhammad's programme of consolidating his 
own strength and building a complex of tribes in alliance with 
himself. It should further be noticed that by agreeing to a pact of 
non-aggression for ten years Muhammad had by implication given 
up the blockade of Mecca. Mecca could presumably now resume 
her trade with Syria, though her monopoly of this trade was gone. 
While flinging away this weapon, however, Muhammad was 
strengthening himself in other ways, and, if need be, could meet 
Quraysh in battle at some future date with good hopes of success. 
Meantime he was reducing his pressure on Mecca and showing 
himself disposed to be friendly and ready to respect Meccan feel- 
ings in various ways^ In other words, instead of vigorously prose- 
cuting the struggle with Mecca, he was angling for the conversion 
of Quraysh to Islam. About what lay behind this aim we cannot 
be quite certain. Perhaps he was merely disgusted at the recent 
refusal of the nomads to join in his pilgrimage, and felt they were 
very unreliable, compared with his fellow tribesmen. But possibly 
he saw that in the new Islamic state their administrative and 
organizing ability would be in demand. Certainly from this time 
onward, whatever may have been the case previously, he was 
aiming at winning the Meccans for Islam and the Islamic state. 
(The believing men and .women for whose sake Mecca was said to 
be spared 1 were doubtless people who did not yet believe but were 
potential believers ; they can hardly have been people who already 
believed but concealed their faith. This is tantamount to saying 
that Muhammad hoped to win many of the Meccans for Islam.) 

The treaty was thus favourable to Muhammad's long-term 
strategy, but for the moment left him to deal with the disappoint- 
ment of his followers at the apparent failure of the expedition. In 
this crisis smouldering embers of dissatisfaction within Muham- 
mad were fanned into flame, and he acted vigorously. He had been 
disappointed when some of the allied nomads refused to join in the 
pilgrimage. They had seen no prospect of booty, and had sus- 
pected that the Muslims might not even return safely. 2 Besides 
making Muhammad's demonstration less impressive, their action 
had shown slight interest in Islam as a religion and little loyalty 
to Muhammad. 

1 Q. 48. 25. * Ibid. 12. 

5783 E 


It is against this background that the Pledge of Good Pleasure 
(bay'at ar-ridwan) must be considered. The usual account is that 
Muhammad eventually sent 'Uthman b. 'Affan to discuss matters 
with the Meccans; as a member of the clan of c Abd Shams he had 
powerful protection in Mecca. When he was long in returning and 
a rumour got about that he had been killed, Muhammad called 
the Muslims to himself under a tree and made them pledge them- 
selves to him. A few sources say that this was a pledge to fight to 
the death (bdycfn 'aid 'l-mawi), but most of them explicitly deny 
that and say it was a pledge not to flee. One account, however, says 
it was a pledge to do whatever Muhammad had in mind. 1 

It is tempting to think that the last gives the essence of the 
pledge. What the situation demanded most of all was that the 
Muslims should accept Muhammad's decision, even if it seemed 
to them unsatisfactory; and they did in fact agree to the renuncia- 
tion of the plan of making the pilgrimage that year. 'Umar is said 
to have protested ; and the unwillingness of most of the Muslims 
to sacrifice the animals and shave their hair after the signing of the 
treaty may be due to a feeling that they had not duly performed 
the rites. Such a pledge to accept Muhammad's decision would 
be an advance on his part towards the position of autocrat; but the 
refusal of many nomads to join the expedition may have made 
him regard some such strengthening of his position as necessary. 
It must have been about this time that he started in suitable 
cases to insist on acceptance of Islam and readiness to obey the 
Messenger of God as conditions for alliance with himself; 

While general considerations point to this interpretation of the 
pledge, it has to be admitted that the evidence is slender. The 
Arabic name, bay' at ar-ridzvdn, could conceivably mean a pledge 
to do what seemed good to Muhammad, but it is almost certainly 
derived from the Qur'anic words: 'God was well-pleased (radiya) 
with the believers when they took the pledge to thee under the 
tree.' 2 On the other hand, the commonest version, namely, that 
it was a pledge not to flee, not a pledge to the death, is compatible 
with the interpretation just suggested. The insistence that it was 
not a pledge to the death is perhaps an indication that it was not 
solely connected with fighting. It is indeed curious that so much 
should be made of the distinction between 'not fleeing' and 'fighting 
to the death', for the latter seems to be involved in the former. It 
1 WW, 254- * Q- 48. 1 8. 


is simplest, therefore, to suppose that the pledge was not an oath 
never under any circumstances to flee, but an oath not to make 
for safety unless with Muhammad's permission. Though this was 
the aspect emphasized at the time, it may well be that the pledge 
was essentially one to accept his judgement in general. The period 
of tension when it looked as if the Meccans would decide to fight 
would be a convenient opportunity for demanding such a pledge. 

Whatever the precise content of the pledge, Muhammad was 
certainly well able to control the Muslims at al-Hudaybiyah, and, 
as the sources indicate, this made a great impression on some of 
the Meccan negotiators. It was natural, however, for an Arab to 
feel that virtue could not be allowed to go unrewarded ; and before 
long perhaps on the way back to Medina Muhammad had 
evolved the scheme .of attacking the rich Jewish settlement of 
Khaybar, but allowing only those who took the pledge at al- 
Hudaybiyah to participate. There were weighty military reasons 
for the attack, but it would also, if successful, reward those who had 
been faithful to him. Some six months later he carried out the plan. 

As he rode home to Medina, Muhammad must have been well 
satisfied with the expedition. In making a treaty with the Meccans 
as an equal he had received public recognition of the position that 
was clearly his after the failure of the siege of Medina. More impor- 
tant was the fact that, by ending the state of war with Mecca, he 
had gained a larger measure of freedom for the work of extending 
the influence of the religious and political organization he had 
formed. He doubtless realized that some of the pagan Meccans 
had been impressed. Yet in stopping the blockade Muhammad had 
made a great military and economic concession, and what he 
had gained in return was chiefly among the imponderabilia. The 
treaty of al-Hudaybiyah was only satisfactory for the Muslims in 
so far as one believed in Islam and its attractive power. Had 
Muhammad not been able to maintain and strengthen his hold on 
the Muslims by the sway of the religious ideas of Islam over their 
imaginations, and had he not been able to attract fresh converts 
to Islam, the treaty of al-Hudaybiyah would not have worked in 
his favour. Material reasons certainly played a large part in the 
conversion of many Arabs to Islam. But any historian who is not 
biased in favour of materialism must also allow as factors of 
supreme importance Muhammad's belief in the message of the 
Qur'an, his belief in the future of Islam as a religious and political 


system, and his unflinching devotion to the task to which, as he 
believed, God had called him. These attitudes of Muhammad 
underlay the policy he followed at al-Hudaybiyah. 

This expedition and treaty mark a new initiative on the part of 
Muhammad. His had been the activity which provoked Quraysh 
after his migration to Medina. Their ripost had failed. The obvious 
way for Muhammad to follow up his advantage would have been 
to set about destroying the influence of Mecca. Instead of that, 
however, he tried something new. 


In the period of nearly two years between the treaty of al- 
Hudaybiyah (March 628) and the capture or surrender of Mecca 
(January 630) some seventeen Muslim expeditions are recorded, 
though of some practically no details are given. 1 These expeditions, 
if that to Khaybar is neglected, may conveniently be regarded as 
falling into three groups. 

Firstly, there are a number of expeditions against tribes which 
were ceasing to oppose Muhammad but were not yet completely 
quiescent. Ghatafan, including Murrah and Tha'labah, attracted 
most attention. A small expedition to Fadak against Murrah led 
by Bashir b. Sa'd was unfortunate, and most of the Muslim party 
were killed; but revenge was taken immediately by a larger expe- 
dition under Ghalib b. 'Abdallah. Thesame Bashir successfully 
led a force of 300 men against Ghatafan two months later. Pre- 
sumably only some sections of Ghatafan were involved, and not 
the whole; but 'Uyaynah the chief, to whom Muhammad had 
made a gift from the spoils of Khaybar, was present and took to 
flight. 2 Another old enemy, Sulaym, was the object of an expedition 
under Ibn Abi 'l-'Awja', himself a Sulami. Though most of the 
fifty participants were killed, there is no mention of revenge being 
taken, so it is probable that this was an old quarrel within the 
tribe which was being prosecuted by one party in the name of 
Islam. Less than a year after this a large contingent of Sulaym 
joined Muhammad for the expedition against Mecca. The expedi- 
tion of Ghalib b. 'Abdallah al-Laythi against a branch of Layth 
called Mulawwih is probably also a repayment of old scores. The 
expedition of Abu 'Ubaydah against some of the Juhaynah in 

1 General sources for the section : IH, 788-802; WW, 297-319; Tab. 1552- 
1618; Caetani, ii/i. 3-105. * Cf. IS, iii/2, 84. 


November 629 (viii/8) is puzzling, as most of Juhaynah were in 
alliance with the Muslims. 

Secondly, there are a number of expeditions against parts of 
Hawazin. These were perhaps of no great importance in them- 
selves ; but they are significant as an indication of the geographical 
expansion of Muhammad's power and as a premonition of the 
events which culminated in the battle of Hunayn. 

Thirdly, there are the expeditions to the north, which may be 
reckoned as three in number, namely, that under Ka'b al-Ghifari 
to Dhat Atlah, that to Mu'tah, and that under f Amr b. al-'As to 
Dhat as-Salasil. Some of the expeditions of the first group may 
also have been concerned with the road to Syria. Certainly Muham- 
mad was intensely interested in the route to the north. The expedi- 
tion to Dumat al-Jan^al in the autumn of 626 l marks an early stage 
in the growth of this interest. There were further expeditions 
towards the end of 627 and the beginning of 628. Now the expedi- 
tion to Mu'tah in September 629 (v/8), led by Muhammad's 
adopted son, Zayd b. Harithah, shows the high importance 
attached to the northern road. Three thousand men took part in 
this expedition nearly twice as many as at al-Hudaybiyah, and 
half as many again as at Khaybar. Muhammad doubtless did not 
tell many of his followers the plans that were in his mind, and 
consequently our sources are exasperatingly silent on events con- 
nected with Muhammad' t s Northern' policy. We do not know, for 
example, on what errand Muhammad had sent Ka'b al-Ghifari 
and his party of fourteen over the border into Syria ; we may be 
almost certain, however, that they were not mere raiders, but were 
carrying out part of some far-reaching plan. The massacre of this 
party by men of Quda'ah would be part of the reason for the two 
later expeditions, since there were some Quda'ah among the enemy 
at Mu'tah. 2 

The great expedition to Mu'tah is not merely part of the 
mysterious 'northern' policy, but is in itself mysterious. A mes- 
senger from Muhammad to the prince of Bostra (Busra) had been 
intercepted and put to death by Shurahbil b. 'Amr of Ghassan, 
and Zayd was sent out with 3,000 men to exact a penalty. Muham- 
mad may have thought of taking part himself, in view of the 
importance and size of the expedition, for it was apparently only 
after the men were collected that Zayd was appointed supreme 
1 Cf. p. 35 above. 2 WW, 314- 


commander; in the end, however, poor health (perhaps the result 
of an attempt to poison him at Khaybar) or pressure of affairs kept 
him at home. The expedition made its way north, and at Ma'an 
heard of a large Byzantine army ahead of them, including many 
Arab tribesmen. They decided, however, to proceed, andatMu'tah 
the two forces met. Zayd and two other prominent Muslims were 
killed. The Muslims are said to have taken to flight, but to have 
been rallied by Thabit b. Aqram of the Ansar and Khalid b. al- 
Walid. After some further fighting (in which according to one 
source the enemy fled) Khalid decided to lead the force back to 
Medina. On reaching Medina the army is said to have been greeted 
with derision presumably because they had retired without taking 
due revenge for the fallen. 

The story of Mu'tah has been greatly manipulated in trans- 
mission, and it is impossible to be certain of more than the barest 
outline of events. The chief source of confusion has been the desire 
to vilify Khalid. Thus the account of how Muhammad appointed 
Ja'far b. Abl Talib and 'Abdallah b. Rawahah to succeed Zayd if 
he fell is probably an invention to support the accusation that 
Khalid unjustifiably assumed supreme command. (Ja'far and 
'Abdallah were perhaps second and third in command of the 
centre.) The story of how Thabit b. Aqram insisted on his taking 
command is primarily an attempt to counter this. Again, the 
emphasis on the hostile reception of the prmy on its homecoming 
is a denigration of Khalid's decision to return to Medina, though 
this was presumably the wisest thing to do in the circumstances. 
The exaggerated reports of the enemy numbers 100,000 men 
may be part of the defence of Khalid's action. When allowance is 
made for this hostility to Khalid, and for the usual glorification 
of the part played by members of one's family, the following points 
seem to remain: (i) there was some sort of an encounter with an 
enemy force; (2) Zayd, Ja e far, and 'Abdallah were killed, but not 
many others; (3) the army returned to Medina under the command 
of Khalid without serious loss. 

Beyond these points there is much uncertainty. It is unlikely 
that the encounter was with the whole of the opposing army. Al- 
Waqidi says only 8 Muslims were killed, but Ibn Hisham adds 
4 other names. This is an incredibly small casualty list, however, 
for a pitched battle between 3,000 men on one side and, say, 20,000 
or 10,000 or even 3,000 on the other unless, indeed, the Muslims 


completely routed the enemy. On the other hand, had there been 
more casualties they would have been mentioned to blacken Khalid, 
if for no other reason ; but in fact the sources do not seem to try 
to conceal Muslim losses. It is possible, then, that the encounter 
was of the nature of a skirmish. It is difficult to conceive a skirmish 
in which the general and two staff officers were killed, but hardly 
anyone else; but, in view of Arab methods of fighting, it is not 
an absolute impossibility. In this encounter the Arabs may well 
have had the best of it; otherwise the losses would have been 
heavier. The decision to return would be dictated not by danger 
from the enemy or by cowardice, but by the length of absence from 
the base and perhaps also by Khalid's ignorance of the precise aims 
of the expedition. (The alleged instructions to Zayd given in the 
sources seem to bf lopg to a later date.) 1 This is practically all that 
can be said. 

While the expedition was doubtless successful in increasing 
respect for Muslim power, the death of the general may have had 
a contrary effect in some quarters. At least in the following month 
(October 629 = vi/8) an expedition was sent under 'Amr b. al-'As, 
another recent convert, against the tribes of Bali and Quda'ah 
which (or part of which) had been on the Byzantine side at Mu'tah, 
and were reported to be preparing a further concentration of men 
against Medina. After reinforcements had come from Medina and 
a dispute about leadership had been settled, the threatening con- 
centration was completely dispersed. 

With his power over a wide area as secure as power could be in 
Arabian conditions, Muhammad was now in a position to march on 
Mecca as soon as he found an occasion for interference. 


Before we consider Muhammad's triumphal entry into Mecca, 
it will be convenient to review the course of the internal politics 
of Mecca since the battle of Badr. The sources give us no more 
than a glimpse of what was happening even if we assume their 
general reliability. In later times a man's descendants would pass 
over in silence or minimize his opposition to Muhammad, while 
the enemies of the man or of his descendants would exaggerate it, 
and thus there is a ground for doubting the sources on this topic 
over and above the normal grounds for doubting sources. The 
1 Ibid. 309. 22-33; cf Caetani, ii/i. 82, n. 2. 


in. 4 

items of information to be considered in this section, however, are 
too varied in their 'tendency* to be explicable as inventions to the 
honour or dishonour of the actors. 

The political changes at this period illustrate the progressive 
disintegration of the clan system at Mecca. That system was still 
the basis of social security, but apart from that it was unimportant. 
Not merely were the old alliances shifting, but it becomes common 
to find members of the same clan on opposite sides. Politics tends 
to be a matter of individuals rather than of clans. 

The political divisions at the time of Badr may be taken as 
reflected in men's approval or disapproval of the policy of Abu 
Jahl (Makhzum). The latter apparently wanted to precipitate the 
battle, whereas, after the caravan was out of danger, Abu Sufyan 
tried to avoid a clash and considered that Aby Jahl rather * fancied 
himself as leader he was presumably entitled to lead in war only 
during Abu Sufyan's absence, since 'Abd Shams was entitled to 
the qiyddah or leadership in war. 1 The following table summarizes 
the statements of al-Waqidi 2 (an asterisk indicates those killed at 





*Abu Jahi 


'Abdallah b. Rabi'ah 

Jumah . 

'Umayr b. Wahb 

*Umayyah b. Khalaf 

Sahm . 

*Munabbih b. al-FIajjaj 

f Abd ad-Dr . 

*an-Nadr b. al-H5rith 

t t 


Suhayl b. 'Amr 


. . 

Nawfal . 

*Tu'aymah b. 'Adi 

al-Harith b. 'Amir ( ?) 

'Abd Shams . 

*'Uqbah b. Abl Mu'ayt 

Abu Sufyan 

*'Utbah b. Rabi'ah 

Shaybah " 


*Zam'ah b. al-Aswad 

I^akim b. liizam 

Zuhrah . 


From this it appears that some clans were divided in their attitude 
to Abu Jahl, and that c Abd Shams was not united behind Abu Sufyan. 
It is noteworthy, too, that the clans of 'Amir and Nawfal which 
did not belong to the old clan-group of the Ahlaf 3 were now 
apparently whole-hearted supporters of Makhzum. On the other 
hand, there are signs that the Ahlaf ('Abd ad-Dar, Makhzum, 
1 WW, 45. 2 Ibid. 41-43, 50-53. 3 M/Mecca t 5-7. 


Jumah, Sahm, 'Adi) are breaking up; e AdI had already gone be- 
cause of a quarrel with Sahm, and now Sahm appears to be going, 
while Jumah is restive. 

The death at Badr of so many of the leaders caused some shifting 
of the balance of power, while the catastrophic change in the 
fortunes of Mecca made men forget for a time the old rivalry 
between the groups associated with Makhzum and 'Abd Shams 
respectively. Al-Waqidi's list of the women taken with them by 
the Meccans on the expedition of Uhud appears to give an indica- 
tion of the leading men and tribes. 1 These women, with two excep- 
tions, were wives of chiefs of clans, and the fact that Abu Sufyan 
and Safwan b. Umayyah took two each suggests that these were 
now the leaders of the rival factions. Abu JahPs son 'Ikrimah was 
apparently not yet oy a level with Safwan. The men of Quraysh 
who took wives were : 

'Abd Shams . . Abu Sufyan (2). 

Jumah . . . Safwan (2). 

Makhzum . . 'Ikrimah (i); al-Harith b. Hisham (i). 

'Abd ad-Dar . Talhah b. Abi Talhah (i). 

Sahm . . . 'Anir b. al-'As (i). 

'Amir . . . Perhaps represented by Khunas bint 

Malik who accompanied her son Abu 

'Uzayr of 'Abd ad-Dar. 

The other five women were from tribes allied with Quraysh. 
Sufyan b. 'Uwayf of B. 'Abd Manat b. Kinanah had both wife and 
daughter-in-law, and so was almost on a level with Abu Sufyan 
and Safwan. Ad-Dughaynah (or ad-Dughunnah) accompanied her 
two sons, perhaps chiefs of al-Harith b. 'Abd Manat b. Kinanah, 
and so of the Ahablsh. 2 The remaining two I have not been able to 
identify ; they may be from Thaqif, unless one is from the clan of 
Nawfal of Quraysh. Safwan's wives were from Thaqif and Kinanah 

Divided counsels made their appearance immediately after the 
battle of Uhud. Safwan's view of the situation was that the Meccans 
ought to be content with the success they had achieved and not 
endanger it, whereas 'Amr b. al-'As and Abu Sufyan thought they 

1 WW, 102 ;cf. IH, 557. 

2 Cf. IH, 245, and note in 'Kritische Aiimerkungen', p. 80; in IH, 852 
(= WW, 364) a Sulami is son of ad-Dughunnah. 


ought to drive home their advantage by attacking Medina. 1 We 
might conjecture that Safwan was afraid that, as Abu Sufyan was 
supreme commander, a successful campaign might redound too 
much to the latter's glory. It is also possible, however, that Abu 
Sufyan was more of a statesman. In opposing Safwan's earlier 
suggestion of cutting down the palm-trees at Medina he may have 
been hoping to win over some of the Medinans (unless the report 
is simply later white-washing of the Umayyads). 2 He also pursued 
a milder course a few months later when the tribe of Lihyan brought 
the Muslims captured at ar-Raji' to Mecca. In its bitterness against 
the Muslims the 'Makhzum group' to which Safwan belonged 
bought these captives and put them to death to avenge their own 
losses at Muslim hands; the clans involved were Nawfal, Jumah, 
Makhzum, e Amir, Zuhrah, and 'Abd ad-D3r. 3 Abu Sufyan had 
nothing to do with this affair, though his son and other members 
of his clan had been killed at Badr ; a confederate of the clan seems 
to have been involved, however. 

The rivalry between Abu Sufyan and Safwan b. Umayyah abated 
sufficiently to allow a united expedition to Badr in April 626 (xi/4) 
and the attempt to capture Medina in April 627 (xi/5), but the 
dissensions between the Meccan leaders led to delay on both 
occasions. Abu Sufyan had the supreme command in view of the 
hereditary privilege of his clan. 'Ikrimah, however, began to come 
into prominence at the siege of Medina, especially in negotiations 
with the Jews; 4 and by the time of Muhammad's expedition to 
al-Hudaybiyah in March 628 (xi/6) we find the triumvirate of 
Safwan b. Umayyah, Suhayl b. 'Amr ('Amir), and 'Ikrimah b. 
Abi Jahl constituting the core of the resistance to the Muslims. 
When Muhammad suggested negotiations, however, the trium- 
virate was divided. 'Ikrimah was against any negotiations, and at 
one point maltreated Muhammad's envoy, but he was opposed by 
Safwan along with al-Harith b. Hisham of Makhzum. 5 'Ikrimah 
was eventually won over, and when it came to the final negotiation 
of the treaty, this was entrusted to the third of the triumvirate, 
Suhayl, assisted by two of his fellow clansmen. In all this there is 
no mention of Abu Sufyan; and mere absence 6 is not enough to 

1 WW, 138, 146, 150. * Ibid. 103. 

3 IH, 638 ff., esp. 645 ; WW, 158-60; Sa'id b. 'Abdallah b. 'Abd Qays b. f Abd 
Wudd (' Amir) is probably a brother of the 'Amr b. 'Abd (Wudd) b. 'Abd Qays 
killed at the Khandaq. 4 WW, 201, 206. 5 Ibid. 250, 253. 

6 Cf. ibid. 323; but contrast IH, 745 and Tab. 1542 f. 


explain this silence, though absence if he were absent would 
help. Perhaps he felt that there was little point in continuing the 
struggle; the old position of Mecca was irretrievably lost; one had 
to accept a decrease in dignity and prosperity and make the best 
of it. He certainly expected Muhammad to be successful at Khay- 
bar, as soon as he heard of the expedition. 1 Perhaps, too, he was 
growing weaker compared with the 'Makhzum group*. His own 
clan was not giving him wholehearted support, for a member of it, 
al-Hakam b. Ab! 'l-'As (father of the caliph Marwan), joined 
'Ikrimah in opposing the negotiations at al-Hudaybiyah, although 
Abu Sufyan, whether present or absent, presumably supported the 
policy of negotiating. 2 Further, he may have known that Muham- 
mad was proposing to marry his daughter, Umm Habibah, and, 
although she had be$n a Muslim for over a dozen years, this may 
have influenced him. 3 

After the first breath of relief at the signing of the treaty of al- 
Hudaybiyah, Mecca must have felt a doomed city. The older men 
and those with vested interests would want to carry on, but the 
younger men must have seen that there was no future for them in 
Mecca. Abu Jandal, a son of the Meccan plenipotentiary, Suhayl b. 
'Amr, is said to have made his way to the Muslim camp to profess 
Islam at the very time when his father was arranging the treaty; 
and in accordance with the terms of the treaty he was handed back 
to his father. 4 The most notable converts were c Amr b. al-'As, 
probably now chief of Sahm, and Khalid b. al-Walid, a prominent 
member of Makhzum, already noted for his military ability. These 
came to Medina in the summer of 629, and were almost immedi- 
ately given a leading place in the Muslim community and put in 
command of expeditions. With them came a third Meccan, 'Uth- 
man b. Talhah ( c Abd ad-Dar), but he was less prominent. There 
were doubtless others, however, not mentioned by Ibn Hisham 
and al-Waqidl. Thus Aban b. Sa'id ('Abd Shams), who had given 
protection (jiwar) to 'Uthman b. ' Affan when he entered Mecca 
as Muhammad's envoy to arrange the treaty, is said to have been 
converted between the expeditions of al-Hudaybiyah and Khay- 
bar; 5 and Jubayr b. al-Mut'im was perhaps converted before the 
fall of Mecca. 6 Presumably many, even if not all, of the 700 

1 WW, 289. 2 Ibid. 250. 

3 Tab. 1571. 10; cf. Caetani, ii/i. 55, &c. 4 WW, 256. 

5 Usd, i. 35 f.; cf. ad-Diyarbakri, Al-Khamts, (Cairo), (i884)/i3O2, ii. 46 f. 

6 Usd, i. 270. 


Emigrants said to be at Hunayn were from Quraysh. 1 At least one 
woman went to Medina in the period in question, for her advent 
raised a problem about the interpretation of the treaty. 2 And two 
cousins of Muhammad's, Abu Sufyan b. al-Harith b. 'Abd al- 
Muttalib (Hashim) 3 and 'Abdallah b. Abl Umayyah (Makhzum), 
the son of 'Atikah bint 'Abd al-Muttalib, joined him while he was 
on his way to Mecca in January 630. 

Al-'Abbas b. c Abd al-Muttalib, Muhammad's uncle, also went 
to meet him on his way to Mecca, along with Makhramah b. Naw- 
fal (Zuhrah), but it is possible that he was converted earlier. As 
the eponymous ancestor of the 'Abbasid dynasty, al-' Abbas had 
to be 'white- washed* by 'Abbasid propagandists and historians, 
and an attempt was made to show that his long residence in Mecca 
was due to his acting as a secret agent for JVluhammad. This is 
hardly credible before Muhammad's pilgrimage in March 629 
(xi/y). Al-'Abbas was a banker and financier, doubtless in a small 
way, though he probably made something out of the siqdyah or right 
of providing water for the pilgrims. He had no importance in the 
affairs of Mecca, and life there cannot have been very comfortable 
for him. Muhammad's marriage to Maymunah was primarily an 
attempt to win al-'Abbas to Muhammad's cause or to seal his 
allegiance to it. Maymunah was sister of the wife of al-'Abbas and 
belonged to a family where matrilineal kinship was important ; by 
marrying her Muhammad was forging,, a strong link with al- 
'Abbas. 4 The fact that another member of the matrilineal family 
(and of the household of al-'Abbas), Hamzah's daughter' 'Am- 
marah or Umamah was taken to Medina at the same time was it 
for safety ? supports the view that al-'Abbas became a Muslim at 
this point and remained in Mecca to work for Muhammad. 

While some Meccans were thus abandoning the ship, the re- 
mainder were trying to make the most of the advantages they derived 
from the treaty. Things did not go smoothly, however, and events 
occurred which at the very least caused some friction. There was 
a clause in the treaty according to which Muhammad was to send 
back anyone of Quraysh who went to Medina without permission 
of his protector. In the case of married women Muhammad refused 
to apply this; perhaps it was because, as a matter of principle, he 

1 WW, 358; IH, 754 f. * WW, 262 f. 

3 To be distinguished from the better-known Abu Sufyan, who is Abu 
Sufyan b. IJarb. 4 Cf. p. 288 below. 


insisted on treating women as independent persons. 1 In the affair 
of Abu Basir, too, he connived at and possibly encouraged what 
looks like a breach of the treaty, though certainly by Arab standards, 
and perhaps even by those of the West, his conduct was formally 
correct; it was not officially questioned by the Meccans. This affair 
is worth describing in detail. 2 

Abu Basir 'Utbah b. Usayd, by origin belonging to Thaqif, but 
now a confederate of the clan of Zuhrah at Mecca, had been 
imprisoned for his Muslim sympathies, but managed to make his 
way to Medina to Muhammad. On his heels, however, came a man 
of the clan of 'Amir bearing a letter from the heads of Zuhrah 
demanding his extradition. Muhammad acknowledged the justice 
of the request, and, when Abu Basir protested, said that God would 
make a way out qf lys difficulties and would not allow him to be 
seduced from his religion. The 'Amiri, his freedman and their 
prisoner had not gone many miles before Abu Basir seized an 
opportunity. When they halted for lunch he won the confidence of 
the others by sharing his dates with them ; they had only dry bread, 
for dates were a Medinan product. The 'Amiri took off his sword 
to be more comfortable, and on Abu Basir's praising it and asking 
if it were sharp unsheathed it and let him put his hand on the 
hilt. It was the work of a minute to kill the unwary captor. The 
freedman escaped to Muhammad, but when Abu Basir also 
appeared and Muhammad gave the freedman the chance of escort- 
ing him back to Mecca, he not surprisingly declined. As Abu Basir 
had been handed over to Quraysh, he was no longer technically 
a Muslim and Muhammad had no responsibility technically for the 
bloodshed. To maintain his correct attitude Muhammad refused 
the fifth of the booty offered to him. 

Quraysh, however, would now be more than ever incensed at 
Abu Basir and could require him from Muhammad ; so with some 
words of encouragement from Muhammad he left Medina and 
went to a spot near the coast which commanded Quraysh's route 
to Syria. Here again probably not without Muhammad's en- 
couragement there gathered round him seventy would-be Mus- 
lims from Mecca, whom Muhammad would have had to hand back 
had they gone to Medina. This band attacked small caravans be- 
longing to Quraysh and killed any man that came into their power. 
In this way, without breaking the letter of the treaty, Muhammad 
1 Cf. p. 283 below. a IH, 751-3; WW, 261 f. 


partly restored the boycott. As these men were not officially mem- 
bers of his community he had no responsibility for their actions. 
Quraysh, on the other hand, though free to use violence on the 
men so far as Muhammad was concerned, were now too weak to 
do so at such a distance from Mecca. In the end they appealed to 
Muhammad to take the men into his community, presumably 
agreeing to waive their rights under the treaty. Abu Basir unfortun- 
ately died just as Muhammad's letter to this effect reached him. 
This incident illustrates how the attraction of the religio-political 
system of Islam outweighed the apparent advantages Quraysh 
received from the treaty. 

The affair which put an end to the peace with Muhammad and 
led to his triumph over Mecca was the plot against his allies the 
Khuza'ah. 1 The prime mover was Nawfal b. Mu'awiyah of ad- 
Du'il, but most of the leading men of the 'Makhzum group* 
supported him. Before the coming of Islam there had been a feud 
between B. Bakr b. ' Abd Manat (a part of Kinanah) and Khuza'ah. 
It was quiescent for a time, but broke out afresh when Khuza'ah 
killed a man of ad-Du'il (a section of Bakr) who had composed 
verses hostile to Muhammad. After al-Hudaybiyah Khuza'ah had 
openly pronounced themselves allies of Muhammad and Bakr of 
Quraysh. Nawfal b. Mu'awiyah secretly got a quantity of weapons 
from the leaders of Quraysh and plotted to take Khuza'ah by sur- 
prise. Things went according to plan ard Khuza'ah, after some 
losses, fled to the houses of two fellow tribesmen in Mecca. The 
Meccan leaders had aided and abetted the plotters. Safwan and 
two men of 'Amir were even said to have been present in disguise, 
but this is probably later calumny, though they were certainly 
privy to the plot. So also were Suhayl and 'Ikrimah. On the other 
hand two important members of Makhzum knew about it but 
thought it unwise al-Harith b. Hisham and 'Abdallah b. Abi 
Rabi'ah. Abu Sufyan was evidently ignorant. 

A man of Khuza'ah reported at once to Muhammad. Quraysh, 
after stopping the fighting and presumably sending the men of 
Bakr out of the city, realized that the situation was serious. If they 
were not to submit to Muhammad, they had a choice between 
three courses: they might disown the section of Bakr involved, 
Nufathah, and let Muhammad do what he liked with them; they 
might pay blood-money ; they might declare war on Muhammad. 
1 IH,8o2ff.; WW, 319 ff. 


Suhayl was for the first option, perhaps influenced by the fact that 
his mother was of Khuza'ah, and also by an old feud between 
'Amir and Bakr. 1 The other courses likewise found some support, 
but there was no agreement about which was best. To pay blood- 
money would mean a great loss of face, whereas a rupture of the 
treaty would lead to economic loss and there was little hope of 
defeating Muhammad. In the end Abu Sufyan persuaded Quraysh 
to attempt a compromise, and he himself was sent to Medina to 
try to secure this. It was a sign of how the mighty Quraysh had 
fallen that they now had to go to Muhammad and ask him humbly 
for a favour. The sources do not give a clear account of the com- 
promise hoped for, but it seems to have been something analogous 
to the position of Abu Baslr an attempt to use against Muhammad 
the principles involved in that case. Quraysh, it seems, were to 
admit that a wrong had been done, but to maintain that they were 
not responsible for it, possibly because the wrong-doers were not 
included in the treaty or because they had acted on their own; they 
were to renew the treaty, however, from now on so as to include 
these people. Unfortunately Muhammad was not prepared to play 
their game, and he was in a much stronger position than Quraysh 
had been with regard to Abu Basir. After Abu Sufyan's mission 
Quraysh was left with the same three or rather four choices as 
before; but Abu Sufyan was now turning to the fourth sub- 
mission to Muhammad. 

The accounts of what happened while Abu Sufyan was in 
Medina are highly coloured. He is said to have gone first to his 
daughter Umm Hablbah, now Muhammad's wife, but she refused 
to let him sit on her bed even, since he was an unbeliever and it 
was used by the prophet. Muhammad himself refused to speak to 
him. He then went in turn to Abu Bakr, 'Umar, Uthman, and 'All 
to ask assistance, but all he got was some advice from 'All. Now, 
while Umm Habibah's lack of filial respect is not inconceivable 
at this period, though unlikely if Muhammad used his marriages 
to win over opponents, the excessive reverence for the person of 
the prophet must belong to a later date ; and the naming of the 
first four caliphs in order is also suspicious. All we can be certain 
about is that Muhammad refused Abu Sufyan's original proposal. 
After that, even if we reject the story of his visiting the four 
caliphs-to-be, it is possible that while still in Medina he made some 

1 WW, 43. 


pronouncement of jiwdr or giving of protection (as 'All is said to 
have advised). It may be, however, that the pronouncement was 
made later. Whatever its date, Abu Sufyan was the first of the 
Meccan leaders to accept the inevitability of submission to Mu- 
hammad, and his pronouncement of jiwdr y whether of B. Nufathah 
or of Quraysh in general, must be linked up with this change of 
attitude. He could not have hoped to protect men against Muham- 
mad. On the other hand, if he had made terms with Muhammad, 
then Muhammad would support his jiwdr. For others to accept the 
jiwdr of a man who had submitted to Muhammad was tantamount 
to submitting to Muhammad ; but for the proud Meccans the bitter 
pill of submission was thereby sugared over. Abu Sufyan's part in 
the Muslim capture of Mecca is much more important than is 
commonly realized. It has probably been deliberately obscured 
by the sources to avoid making his role appear more glorious than 
that of al- f Abbas. 

What Abu Sufyan did in the last critical days, however, when 
Muhammad with a large army was nearing Mecca, could not be 
concealed. The 'Makhzum group' under Safwan, 'Ikrimah, and 
Suhayl were trying to organize some resistance. Abu Sufyan, on 
the other hand, went out to meet Muhammad, accompanied by 
Hakim b. Hizam (Asad) and Budayl b. Warqa', who, though 
belonging to Khuza'ah, had a house in Mecca. This seems to 
indicate that those outside the 'Makhzijm group* had decided to 
capitulate. Muhammad showed himself eager to avoid bloodshed, 
and acknowledged Abu Sufyan's jiwdr by ordering that those who 
took refuge in Abu Sufyan's house or closed their own houses 
would be safe. In this way the resistance was greatly reduced. The 
ground had doubtless been carefully prepared by Abu Sufyan. 
The leaders only of the 'Makhzum group* fought, along with one or 
two others, notably some tribesmen of Hudhayl. Most of the group 
took advantage of the amnesty gained by Abu Sufyan. 

Abu Sufyan seems to have had a more statesmanlike grasp of 
realities than his Meccan opponents and, after the failure of the 
great confederacy at the siege of Medina, to have seen the hope- 
lessness of continued resistance. He was probably reconciled to 
a decrease in dignity and importance and to a lower standard of 
living. His influence sems to have been on the side of modera- 
tion. Both before and after the siege we find him fostering unity 
among Quraysh and trying to prevent internal strife. It is not 


surprising that in the period after the siege we hear less of him 
than of his more vociferous opponents. The dignity and honour 
of their clans, their vested interests, and some sheer pigheadedness 
made these resist after resistance was hopeless. When the force of 
Muslim arms pushed them out of Mecca, they fled in various 
directions and went into hiding. Most were at length reconciled 
to Muhammad and swallowed their pride, but perhaps most 
typical of their defiant spirit was Hubayrah b. Abi Wahb (Makh- 
zum) who remained in Najran for the rest of his life. 1 


The size and consequent importance of the expedition to Mu'tah 
is an indication that Muhammad's strategic aim by 629/8 was the 
unification of the Arats under himself and their expansion north- 
ward. The capture of Mecca was therefore not an end in itself. 
Nevertheless Mecca and the Meccans were important for Muham- 
mad. Mecca had long since been chosen as the geographical focus 
of Islam, and it was therefore necessary that the Muslims should 
have complete freedom of access to it. Could Mecca be brought 
under his sway, his prestige and power would be greatly increased; 
without Mecca his position was comparatively weak. Moreover, 
as the affairs of the Islamic community grew in volume, Muhammad 
had need of the military and administrative abilities of the Meccans. 
Sooner or later he must try to get Mecca on his side. 

In the year 628 at al-Hudaybiyah it had suited Muhammad to 
make peace and end the blockade, for he was then able to devote 
greater energy to the work among the nomadic tribes. In the 
twenty-two months following the treaty, however, his strength 
grew rapidly; and when his allies of Khuza'ah appealed for help 
he apparently felt that the moment had come for action. If he 
was still uncertain, Abu Sufyan's visit to Medina presumably 
made him realize that few in Mecca would now resist and that 
the 'diehard' leaders of the 'Makhzum group' would have little 
support. He therefore set about collecting a force sufficient to 
overawe the Meccans and ensure that none but the most inveterate 
opponents resisted actively. 

During the preparations for the expedition Muhammad took 
precautions to secure a large measure of secrecy. Nothing was said 

1 Caetani, ii/i. 134. * IH, 802-40; WW, 

5783 F 


in Medina about the goal of the expedition, a small party was sent 
towards Syria to put men on a false scent, and the roads to Mecca 
were sealed off. By a strange lapse (which he alleged to be due to 
anxiety about wife and children in Mecca) one of the veterans of 
Badr tried to give information to Quraysh, but his letter was inter- 
cepted. Muhammad's messengers to the various allied tribes were 
successful, and on i January 630 (io/ix/8) he was able to set out 
with an army which, including those who joined en route, numbered 
about 10,000 men. This included contingents from the tribes: 
1,000 men from Muzaynah, 1,000 (or 700) from Sulaym, 400 each 
from Aslam and Ghifar, and unspecified numbers from Juhaynah, 
Ashja', Khuza'ah (sc. Ka'b), Damrah, Layth, and Sa'd b. Bakr; 
there were also small groups from Tamim, Qays, and Asad. 1 

In due course the army encamped at Man; a2;-Zahran, two short 
stages from Mecca, but still on the road an army would take if 
making for at-Ta'if or the country of Hawazin. The Meccans were 
thus not quite certain of the destination of the army, and indeed 
had probably received little exact information about it. Ten thou- 
sand fires lit by Muhammad's orders increased their dismay. To 
this camp Abu Sufyan came to make his submission; and from it 
he went back to Mecca with word of the general amnesty. The 
following night Muhammad pitched camp nearer to Mecca at 
Dhu Tuwa; in the morning his forces, divided into four columns, 
advanced into Mecca from four directions. Only one column, that 
under Khalid, met with resistance, and that was soon overcome. 
After twenty-four men of Quraysh and four of Hudhayl had been 
killed, the rest fled. Two Muslims were killed when they mistook 
their way and ran into a body of the enemy. With such negligible 
bloodshed did Muhammad achieve this great triumph. 2 The date 
was probably about n January 630 (zo/ix/8). 3 

This event came to be known as the Fat'h or Conquest par 
excellence. 4 The word fat 9 h properly means 'opening', but it is also 
used in other ways, for example of God's bestowing gifts, especially 
rain, on men. The phrase fataha bayna-hum, literally 'he opened 
between them', means 'he judged between them', 5 and so the 
noun fat 9 h comes to be 'used in the sense of something which will 
clear up a doubtful situation'. 6 In late Meccan and early Medinan 

' IH, 810, 828; WW, 326, 332; cf. 358. a IH, 810-18; WW, 330-5. 

3 IH, 840; WW, 355; cf. IH, 810 and WW, 330, but contrast WW, 350. 

4 Cf. Q. 57. 10. * Lane, s.v. 
6 Bell, Translation, note on no. i. 


days Muhammad and his followers seem to have expected a fat 9 h, 
a decision between themselves and the pagans, or perhaps *a de- 
cisive clearing away of the clouds of opposition and distress which 
surrounded' them, and the Qur'an has to meet the objection that 
the/tfJ'A was long in coming. 1 With the victory of Badr it was said 
to have come. 2 But the conception was a wide one and apparently 
could also apply to the signing of the treaty of al-Hudaybiyah. 3 
Muhammad's triumphal entry into Mecca, however, was the final 
and absolute decision between the Muslims and their chief oppo- 
nents, the pagan Quraysh, and as such it came to be regarded as the 
supreme fa? A, though the Qur'anic basis for this use of the term is 
slender. 4 Since this event was also a victory and a conquest, the 
word was used by the next generation of Muslims to describe their 
overrunning of th$ Rersian and Byzantine empires. The meaning 
of conquest, however, is derived from this conception of the con- 
quest of Mecca as a judgement or clearing-up. 

The pursuit of the fleeing pagans was not energetic seeing that 
Muhammad had proclaimed a general amnesty. 'Abdallah b. Abi 
Rabf ah (or Zuhayr b. Abi Umayyah) and al-Harith b. Hisham, 
both of Makhzum but critics of the attack on Khuza'ah, fled to 
the house of a fellow clansman, Hubayrah b. Abi Wahb, whose 
wife was a daughter of Abu Talib and so a cousin of Muhammad's. 
Suhayl b. c Amr went to his own house and sent a son to ask for 
security. His friend Huwaytib was found and assured that all was 
well by the celebrated Abu Dharr, who apparently knew him. 
Safwan b. Umayyah evidently judged it wise to flee to the Red Sea 
coast, but a member of his clan of Jumah obtained an explicit 
guarantee of his security from Muhammad and communicated this 
to him. Muhammad's policy of forbidding all pillage meant that 
some of his poorer followers were now in want, and from some of 
the rich men of Mecca whom he had treated so magnanimously 
Muhammad requested loans. Safwan is said to have lent 50,000 
dirhams, and 'Abdallah b. Abi Rabi'ah and Huwaytib 40,000 each; 
from this the men in need received 50 dirhams apiece. 5 These 

1 Q. 32. 28-30; Bell, op. cit., note on 32. 9/8. 

2 8. 19, and probably 48. i originally. 

3 48. i ; cf. Bell's interpretation, also remarks in E. M. Wherry, A Comprehen- 
sive Commentary on the Quran; comprising Sale's Translation and Preliminary 
Discourse with Additional Notes and Emendations, Boston, 1882-6, ad loc* 

4 Cf. 57. 10 as commonly interpreted; but contrast Bell, ad loc. 

5 IH, 820,825 f.;WW, 336, 343-8- 


leaders were not forced to become Muslims; they and doubtless 
many others remained pagan at least till after al-Ji'ranah. 

There was also a small number of persons specified by name as 
excluded from the general amnesty. Apparently only one of the 
active leaders of the resistance to Muhammad was on this list, 
'Ikrimah b. Abl Jahl, though it is not clear why he was given this 
prominence. His wife, however, after submitting to Muhammad, 
begged pardon for him and obtained it, and had an adventurous 
journey to the Yemen to find him and bring him back. He did not 
receive a gift in the distribution of the spoil at al-Ji'ranah, but this 
may be because he had not yet returned to Mecca at that time. 

The other proscribed persons were all guilty of specific faults. 
Ibn Abl Sarh ('Abdallah b. Sa'd b. Abl Sarh) had been one of the 
Emigrants at Medina and had acted as Muham/nad's amanuensis. 
When Muhammad dictated a phrase of the Qur'an such as sami ' 
'alim, l Hearing, Knowing' (with reference to God), he had written, 
for example, 'alim hakim, 'Knowing, Wise', and Muhammad had 
not noticed the change; he had therefore doubted the reality of 
Muhammad's inspiration, become an apostate, and gone to Mecca. 
Muhammad pardoned him, after some hesitation, on the inter- 
cession of 'Uthman. 'Abdallah (b. Hilal) b. al-Khatal, of one of the 
lesser clans of Quraysh, had been sent out from Medina by Mu- 
hammad to collect sadaqah or legal alms; he became so annoyed 
with the deficiencies of his servant, also a Muslim, that he gave 
him a beating from which he died; he then made for Mecca, taking 
with him the money he had collected. Worse than this he composed 
verses satirizing Muhammad, and these were sung in public by 
two singing-girls of his who were now also proscribed; he himself 
and one of the girls were executed, but the other was pardoned. 
Another singing-girl, Sarah, was also proscribed; according to Ibn 
Hisham she was eventually pardoned, but al-Waqidi says she was 
executed, since, after being pardoned, she repeated the offence. 1 

Besides these spreaders of anti-Muslim propaganda, some perpe- 
trators of acts of violence were put to death where they had contra- 
vened the basic principles of social security in the Islamic state or 
had attacked women of Muhammad's family. Miqyas b. Dubabah 
(or Subabah) al-Laythi had been a Muslim; when his brother was 
killed by mistake on the expedition to Muraysf in January 627 
(viii/5), he had accepted the blood-money paid by Muhammad, and 
' IH, 820; WW, 347- 


the affair ought to have been settled; but when the opportunity 
came, he killed the man responsible and fled to Mecca; he was now 
executed. Al-Huwayrith b. Nuqaydh of B. *Abd of Quraysh was 
executed for knocking down Muhammad's daughters when al- 
' Abbas was taking them to Medina. Habbar b. al-Aswad, guilty of 
similar conduct which had caused Zaynab to have a miscarriage, 
managed later at Medina to appear before Muhammad and, before 
the latter had time to order his punishment, to repeat the shahddah 
or profession of faith; this made him a Muslim, and he was par- 
doned. Hind, the wife of Abu Sufyan, perhaps proscribed because 
she instigated Wahshi to kill Hamzah, also appeared before Mu- 
hammad and made her submission, which was accepted. Thus in 
the end very few persons were put to death. 

Muhammad regained fifteen or twenty days in Mecca. The 
Ka'bah and the private houses were cleansed of idols. Parties were 
sent to destroy Manat at Mushallal (between Mecca and Medina), 
'Uzza at Nakhlah, and various others. A number of pressing 
administrative matters were dealt with, especially the defining of 
the boundaries of the sacred territory of Mecca. Most of the old 
offices or privileges of Quraysh were abolished, but 'Uthman b. 
Talhah of 'Abd ad-Dar retained the custody of the Ka'bah and 
al-' Abbas the right of supplying water to pilgrims. 

Foremost among the reasons for this success of Muhammad's 
was the attractiveness of Islam and its relevance as a religious and 
social system to the religious and social needs of the Arabs. In 
Muhammad in Mecca an attempt was made to analyse the malaise 
of the times, and its root was traced to the transition from a nomadic 
to a settled economy. The Meccan leaders adhered to the old tribal 
standards and customs when it was to their advantage ; but those 
who were not leaders were chiefly aware of the disadvantages. As 
hardships multiplied through the Muslim blockade, the private 
interests of the leaders would come more and more into conflict 
with one another, and unity became more and more difficult to 
preserve. Abu Sufyan probably saw more clearly than the others 
the need for unity among Quraysh, and as hopes of this faded he 
must also have been aware that by going over to Muhammad 
before the last possible moment he would probably strengthen his 
position relatively to the 'Makhzum group'. 

Again, Muhammad's own tact, diplomacy, and administrative 
skill contributed greatly. His marriages to Maymunah and Umm 


IJablbah would help to win over al-'Abbas and Abu Sufyan, and 
he probably gained advantages from the discord at Mecca of which 
we are not aware. Above all, however, his consummate skill in 
handling the confederacy he now ruled, and making all but an 
insignificant minority feel they were being fairly treated, height- 
ened the contrast between the feeling of harmony, satisfaction and 
zest in the Islamic community and the malaise elsewhere; this 
must have been obvious to many and have attracted them to 

In all this one cannot but be impressed by Muhammad's faith 
in his cause, his vision and his far-seeing wisdom. While his com- 
munity was still small and devoting all its energies to avoiding 
being overwhelmed by its enemies, he had conceived a united 
Arabia directed outwards, in which the Mecoan$ would play a new 
role a role no less important than their old role of merchants. He 
had harried them and provoked them ; then he had wooed them and 
frightened them in turn ; and now practically all of them, even the 
greatest, had submitted to him. Against considerable odds, often 
with narrow margins, but nearly always with sureness of touch, 
he had moved towards his goal. If we were not convinced of the 
historicity of these things, few would credit that a despised Meccan 
prophet could re-enter his city as a triumphant conqueror. 


During the time he spent in Mecca Muhammad sent out at 
least three small expeditions to secure the submission of tribes 
in the surrounding district. These expeditions were presumably 
successful, but few details have been preserved except about the 
third, that of Khalid against B. Jadhimah (of Kmanah). The stan- 
dard account of this expedition, however, is hardly more than a 
circumstantial denigration of Khalid, and yields little solid his- 
torical fact. It is not surprising that there are few memories of these 
events, for much else was happening at the time. The Muslims, 
especially the Emigrants, had to adjust their feelings to the sudden 
change whereby their bitterest enemies had become allies. Mu- 
hammad himself had taken over the responsibility for the adminis- 
tration of Mecca. Above all there was a serious military threat in 
that Hawazin and Thaqlf were collecting an army twice the size 
of Muhammad's only two or three days' march away. 

For an understanding of the campaign of IJunayn it is important 


to realize that Hawazin and Thaqlf were old enemies of Quraysh; 
during Muhammad's lifetime there had been fierce fighting on 
several occasions. This had been connected with the trade rivalry 
between Mecca and at-Ta'if (the city of B. Thaqif). The trade of 
at-Ta'if had come under the control of the Meccan merchants, 
who worked through one of the two political groups into which 
Thaqlf were divided, the Ahlaf. The decline in the prestige of 
Quraysh must have upset the balance of power in at-Ta'if and 
given the upper hand to the other group, B. Malik, so that in 
January 630 the city as a whole joined Hawazin against Quraysh. 
This helps to explain why B. Malik fought stubbornly at Hunayn 
and lost nearly a hundred men, whereas the Ahlaf fled almost at 
once and lost only two. 1 While many of Thaqlf obviously hoped to 
assert their independence of Quraysh, the precise expectations of 
Hawazin are obscure. They are said to have started concentrating 
as soon as they heard of Muhammad's preparations at Medina, 
and this may indicate that they regarded the expansion of Muham- 
mad's power as a threat to themselves, though, in view of his policy 
of uniting the Arabs, it is unlikely that he planned to attack them. 
The weakness of Mecca may have made them think they could 
bring it into subjection, but probably they only hoped to pay off 
old scores. A conflict between Muhammad and the Meccans must 
have seemed inevitable, and the presumed exhaustion of both 
sides would give an advantageous opportunity for attacking one 
or both. 

On the Meccan side there was realization of this danger. At no 
point in the sources is there any suggestion that the leaders of the 
Meccan resistance to Muhammad sought help from Hawazin and 
Thaqlf. Even when they fled none of them, so far as we know, 
went in this direction. Feeling between these tribes and Quraysh 
must have been strong. In this situation Muhammad, on becoming 
conqueror of Mecca, at once became also its champion against the 
threatening enemy. It was self-preservation rather than hope of 
booty that made the pagan Quraysh go out with him to Hunayn. 
Safwan b. Umayyah thought submission to Muhammad preferable 
to subjection to Thaqlf or Hawazin; 2 and he lent arms to Muham- 
mad as well as the money mentioned above. Altogether Muhammad 
was able to add 2,000 men to his army, and judged himself strong 

1 WW, 362 ; for the tribal relations cf. p. 101 below. 
MH,8 4 s;WW, 357, 363. 


enough to march out and give battle to an enemy reputed to have 
a force of 20,000. 

Muhammad left Mecca on 27 January 630 (6/x/8) and on the 
evening of the 30th camped at Hunayn close to the enemy. The 
next morning the Muslims moved forward down Wadi Hunayn in 
battle order; the vanguard, commanded by Khalid b. al-Walid, 
included many men of Sulaym. The Muslims, who had been over- 
confident, 1 were somewhat dismayed at the huge mass of human 
beings and animals which they saw, for Hawazin had brought all 
their women, children, and livestock, staking everything on the 
issue of the battle. Suddenly the enemy cavalry, posted overnight 
in the side valleys, attacked the Muslim van. Sulaym, though later 
they protested that they fought bravely, 2 are said to have fled 
almost at once, and their consternation affected a large part of 
Muhammad's army. He himself stood firm, however, with a small 
body of Emigrants and Ansar. This turned the tide, and before 
long the enemy were in full flight. Some Thaqif fought bravely for 
a time, then fled to the safety of their walls. The chief of the con- 
federacy, Malik b. ' Awf, with his own tribe of Nasr, held a pass 
to gain time for those on foot; and there seems to have been 
another stand in front of the enemy camp. In the end, however, 
all eiforts proved unavailing. The fighting men were dispersed 
o'r taken prisoner or killed; the women, children, animals, and 
goods fell into the hands of the Muslims t . 

In the battle of Hunayn a larger number of men were involved 
than in any of Muhammad's previous battles, with the possible 
exception of Mu'tah. It does not appear, however, to have been 
a stubbornly fought battle. The names of only four or five Muslim 
dead have been recorded, but these are all men with homes in 
Medina, and there must have been some loss perhaps a consider- 
able loss among Muhammad's nomadic allies. 3 This suggests that 
there was little hand-to-hand fighting. The victory, none the less, 
was notable and important. Hunayn was the major encounter during 
Muhammad's lifetime between the Muslims and the nomadic tribes. 
The collection and concentration of 20,000 men was a notable feat 
for a nomadic chief, and after Malik b. 'Awf's discomfiture none 
cared to repeat it against Muhammad. Instead, so long as Muham- 
mad lived, and particularly in the year 9 of the Hijrah (April 630- 

1 Cf. Q. 9. 25. 2 IH, 850 f. 

3 IH, 857; WW, 368; Tab. 1669; cf. Caetani, ii/i. 166 f. 


April 631), deputations came to Medina from all over Arabia to 
make agreements and alliances. 


From Hunayn Muhammad went on at once to at-Ta'if and set 
about besieging it. He had some siege-engines, probably adopted 
from the Byzantines, but even with these he made little headway. 
After some fifteen days he decided to abandon the siege. Thaqif 
were resisting bravely and there had been casualties among the 
Muslims. If he allowed the siege to drag on, his men would become 
restive, blood would be shed and a final reconciliation with Thaqif 
would be rendered more difficult; in the course of a long siege, 
too, much of the prestige gained at Hunayn would be dissipated. 
Besides he had Hawazin and the booty of Hunayn to attend to. 
Thus, by abandoning the siege he lost nothing of importance, for 
he had other ways of influencing Thaqif in his favour. Nevertheless 
he may have been disappointed or annoyed; this at any rate is 
a possible reason for his sharp treatment of a man who acciden- 
tally kicked him while they were riding back. 

The booty had been left at al-Ji f ranah, not far from Hunayn, 
under the charge of Mas'ud b. f Amr al-Ghifari. The prisoners 
were there also except that a few of the women had been given to 
the leading Companions. There was sufficient booty to give every 
man in the Muslim army four camels or the equivalent. There is 
said to have been trouble over the distribution and complaints at 
the delay. Some of the new recruits tried to keep pieces of booty 
they had picked up, while Muslims of longer standing scrupulously 
returned everything. Such stories, however, seem to be told mainly 
for the edification of the hearers, and may have been invented or 
developed long after Muhammad's time. 1 

To some of the leading men among Qurajsh and the recently 
allied tribes Muhammad gave presents (either from the fifth or 
from the surplus). 2 The list is interesting as showing the importance 
of the various men. 3 


Abu Sufyan 'Abd Shams 100 (300) 

Yazid b. Abl Sufyan 100 

Mu'awiyah b. Abi Sufyn . . . . 100 

1 IH, 880; WW, 366. 2 WW, 376. 

3 IH, 880-2; WW, 375 f.; Tab. 1679-81; cf. also the longer list, apparently 
not given by Ibn Is'haq, but derived independently from az-Zuhr! by Ibn 
Bisham, IH, 882 f. 



Qaklm b. HizSm 
an-Nu^ayr b. al-Hrith 
Usayd b. tfarithah . 
al-'Ala' b. Jariyah 
Makhramah b. Nawfal 
al-tfarith b. Hisham . 
Sa'Id b. Yarbu' . 
afwan b. Umayyah . 
'Uthman ('Umayr) b. Wahb 
Qays b. 'Adi 
Suhayl b. 'Amr . 
IJuwayfib b. 'Abd al-'Uzza 
Hisham b. 'Amr 
al-Aqra* b. liabis 
'Uyaynah b. rjin 
al- 'Abbas b. Mirdas . 
(Malik b. 'Awf . 


'Abd ad-DSr 


Zuhrah (falif) 







in. 7 

100 (?30o) 

100 (50) 



loo (or more) 
50 (100) 




50 (100) 

These men came to be known as al-mif allafah qulubu-hum, 
'those whose hearts are (or are to be) conciliated or united 1 . This is 
commonly taken to mean that the persons in question have just 
become Muslims and have to be strengthened in their attachment 
to Islam, or that they have to be induced to profess Islam. Pre- 
sumably it was at al-Ji c ranah that most, if not all, of those in the 
list made their acknowledgement of Muhammad as Messenger of 
God. The application of the name to all in the list, however, seems 
to be the work of political opponents during the Umayyad period. 
The Qur'anic expression originally referred to a different set of 
people, since it occurs in a prescription for ,the use of sadaqdt or legal 
alms. In the case of the leaders of Quraysh the gifts had probably 
less to do with their conversion to Islam than Muhammad's 
championship of the cause of Mecca against its enemies, Hawazin 
and Thaqif, while those with experience of leadership must have 
admired his skilful handling of difficult situations. 

A hundred camels was probably regarded as a proper share for 
each of the leaders of Muhammad's non-Muslim allies, in view of 
the fact that Muhammad received the Fifth. Those in the list who 
did not come into this category of allied leaders, notably Malik b. 
'Awf and perhaps 'Uyaynah b. Hisn and al-Aqra f , may have 
received gifts to reconcile their hearts. It is not impossible (though 
the sources are silent) that the leading men of the Aws, the Khazraj, 
and the Muslim tribes also received gifts; but they were in a 
different position from men like Abu Sufyan and Safwan, since 
the latter were technically allies, whereas the Muslims were 
directly under Muhammad's command. As Abu Sufyan and IJakim 


b. Ilizam apparently received 300 each (for the mention of Abu 
Sufyan's sons is doubtless a device to conceal the favour shown to 
him by Muhammad), whereas the others received at most 100, past 
services to Islam seem to have been rewarded, namely, their contri- 
bution to the peaceful surrender of Mecca. 

While the siege of at-Ta'if went on, Muhammad was negotiating 
with Hawazin, and about the time of the division of the spoil at 
al-Ji'ranah Malik b. 'Awf and Hawazin decided to accept Islam, 
but asked to have their women and children back. Ibn Ishaq places 
the restoration of the women before the distribution of the animals, 
but it seems more likely that al-Waqidi is correct in placing these 
events in the opposite order. The knowledge that the camels and 
sheep had been divided and that it was the turn of the women next 
would make Hawazii\more eager to achieve a settlement. As it was, 
the restoration of the women was treated as a favour, not as some- 
thing to which Hawazin had a right, nor even which Muhammad 
could command ; and they seem to have made a payment in return. 1 

Muhammad remained at al-Ji'ranah from 24 February to 
9 March 630 (5 to i8/xi/8). From there he set out to make the lesser 
pilgrimage ('umrah) at Mecca, then returned to Medina. In charge 
of his affairs in Mecca he left a young man of the clan of Umayyah 
b. 'Abd Shams, 'Attab b. Asid. His youth he was under thirty 
suggests that the functions cannot have been of first importance; 
but as he kept the position until his death in 634/13, he must have 
discharged his duties efficiently. The fact that he was of Abu 
Sufyan's clan shows that Muhammad, though on good terms with 
most people in Mecca, tended to support Abu Sufyan and not the 
triumvirate, Safwan, 'Ikrimah, and Suhayl. 

For the rest of Muhammad's life we hear practically nothing 
about Mecca directly, though something can be deduced from 
notices about his late opponents. Abu Sufyan helped with the 
destruction of the idol of al-Lat at at-Ta'if. 2 Muhammad made 
him governor of Najran, or part of it, and he is said to have been 
at the battle of the Yarmuk and to have lived on till about 652/32.3 
'Ikrimah was put in charge of the sadaqdt of Hawazin in 630/9, 
and in the revolt after Muhammad's death known as the Riddah 
commanded a loyalist army in 'Uman; he died as a 'martyr' in the 
fighting in Syria. He apparently showed great zeal for Islam, and 

. l Cf. p. ioi below. * IH, 917 f; WW, 384 f. 

3 Usd, v. 216; iii. 12 f.; cf. al-Baladhuri, Futiih al-Bulddn, Leiden, 1866, 59. 


remarks like the following are attributed to him: 'whatever money 
I spent fighting against you, I shall spend as much in the way of 
God'; "I risked my life for al-Lat and al-'Uzza; shall I hold back 
from risking it for God ?' Others mentioned as being killed in the 
fighting in Syria in 636-8 (15-17) are al-Harith b. Hisham (Makh- 
zum) and an-Nudayr b. al-Harith ('Abd ad-Dar). 1 

On the other hand, Safwan b. Umayyah seems to have remained 
in Mecca and to have died there as an old man. 2 Suhayl b. 'Amr 
was apparently still in Mecca on Muhammad's death, for he is 
credited with being the man chiefly responsible for keeping the 
Meccans loyal when there were signs of disaffection in some of the 
tribes and the 'governor* of Mecca did not give a lead. Later, 
however, he also went to Syria. He had the reputation of being 
the most pious of the group of leaders who, came over to Islam 
after the conquest of Mecca; he had engaged in the religious 
exercise of tahannuth in his pagan days. 3 Various others of the 
group do not seem to have left the Hijaz, but probably settled in 
Medina rather than Mecca; such were Makhramah b. Nawfal, 
Sa'id b. Yarbu', Huwaytib b. 'Abd al-'Uzza, and Hakim b. Hizam. 4 

From such details it is clear that Mecca did not recover its 
position as a trading centre. 5 The increase of security over a wide 
area was advantageous to trade, but the new restrictions, like that 
on usury, imposed by Muhammad stopped the old lucrative specu- 
lations in high finance. For the younger and more adaptable men 
war and administration gave a better promise of a career than com- 
merce. Almost all the people of substance had left Mecca: and 
even for commerce Medina was now a better centre. 6 

With the events at al-Ji'ranah there ends what is perhaps the 
most brilliant phase of Muhammad's career. With negligible excep- 
tions (like Hubayrah) the men who a few months before had been 
implacable enemies had now come over to his side. They were 
ready not merely to become his partners, as they had been at 
Hunayn, but to acknowledge him as prophet; and this implied an 
acknowledgement of him as political superior, as chief of the 

Usd t iv. 5 (cf. WW, 345); i. 35 x f.; v. 20 f. 
Ibid. iii. 22 f. 

Ibid. ii. 371-3 ; cf. F. Wiistenfeld, Die Chroniken der Stadt Mekka, Leipzig, 
1858-61, iv. 118. 

Usd, iv. 337 f.; ii. 316 f., 75, 40-42. 
Cf. H. Lammens, art. 'Mekka' in El (i). 
Cf. Usd, i. 352; WW, 61 f. 


'super-tribe* to which they now belonged, without autocratic 
power indeed, but with various privileges which placed him 
above other men. Though some kept themselves to themselves, 
at least a few became enthusiasts for the propagation of their 
new faith. 



THE previous three chapters have surveyed in chronological 
order the course of Muhammad's relations with the Meccans. 
The remaining two years or so of his life were occupied with 
extending his sway over some of the other tribes of the Arabian 
peninsula. These two years might also be dealt with chronologically. 
It will be more satisfactory, however, to divide up the tribes 
geographically, and, taking each tribe or group of tribes separately, 
to consider Muhammad's relations to it botk before and after the 
conquest of Mecca. In this way we shall be able to form an idea 
of Muhammad's policy towards the tribes. 

In the course of the present chapter the complexity of the tribal 
organization will become apparent. 1 The words * tribe' and 'clan' 
will be used, but the application of them will be arbitrary. Ghataf an 
will be called a 'tribe', but Fazarah, which is a part of Ghataf an, will 
also be called a 'tribe'. What one finds is a bewildering multiplicity 
of groups within groups. The Arabs have half a dozen or more 
words for groups of different sizes, which later writers arranged in 
precise hierarchical order; but the usual practice was to call any 
group, whatever its size, 'Banu Fulan', 'the sons of so-and;so' (by 
European scholars sometimes contracted to 'B.', and sometimes 
omitted). Where the group was not known by the name of a real 
or supposed ancestor, but by some descriptive name, 'Banu' was 
inappropriate and not used ; thus a section of the tribe of Thaqlf 
was known as the Ahlaf or Confederates. Sometimes the same 
name is used for both a smaller and a larger group ; thus a section 
of the tribe of Khuza'ah became large enough to form itself into 
the separate tribe of Aslam, but the remainder continued to be 
known as Khuza'ah; therefore in one sense Aslam is a part of 
Khuza'ah and in another sense it is distinct from it. A further 
source of confusion, especially with the smaller groups, is that 
different groups have the same name; Banu Ka'b, for instance, 

1 Cf. W. Caskel, 'The Beduinization of Arabia', in Studies in Islamic Cultural 
History, ed. by G. E. von Grunebaum, Wisconsin (The American Anthropologist), 
i954 36-46. 


may be B. Ka'b b. 'Amr of Khuza'ah or B. Ka'b b. Rabi'ah of 
Hawazin, or even B. Ka'b. b. Lu'ayy of Quraysh, but the sources 
may simply say C B. Ka'b' and assume that the reader knows which 
is intended. 

The division and subdivision of the tribes is not just a matter 
of nomenclature but an important political fact. Within each group 
there were smaller groups intensely jealous of one another, and 
usually pursuing contrary policies. When we hear of a deputation 
from a tribe going to Muhammad, the probability is that it repre- 
sented only one faction in the tribe. Muhammad must have had 
extensive knowledge of the internal politics of each group, and 
showed wisdom in deciding which faction to support. What has 
to be borne in mind is that he was always dealing with a very 
complex situation., , 

The traditional view of the last two years of Muhammad's life 
is that during these two years most of the tribes of Arabia were 
converted to Islam. In particular, the year 9 of the Islamic era 
(April 63O-April 631) is known as 'the year of deputations'. Each 
tribe is supposed to have sent its wafd or 'deputation' to Muham- 
mad, which professed Islam on behalf of the tribe ; arrangements 
were made for giving instruction. Soon, however, some of the 
tribes became restive under the Islamic dispensation ; they specially 
objected to the contributions they had to make to Medina, whether 
in money or in kind. As Muhammad returned to Medina from the 
'farewell pilgrimage' to Mecca in March 632 (xii/io) he was seen 
to be in poor health, and rumours spread. False prophets appeared 
as leaders of revolt against the Islamic state, first al-Aswad in the 
Yemen and Musaylimah in the Yamamah, and then Tulayhah 
among the tribe of Asad. 1 As his health continued to deteriorate 
(though he was still able to attend to business), disaffection grew. 
His death on 8 June 632 (i3/iii/n) led to the outbreak of a series 
of rebellions in various quarters of Arabia. These are regarded as 
primarily religious, and are known collectively as the Riddah or 
'apostasy'. The use of the Arabic name is convenient, but must 
not be taken to involve unquestioning acceptance of the underlying 
historical conception. 

In opposition to this traditional view some European scholars 
have held that the extent of conversion has been greatly exagger- 
ated, and that only a few tribes round Medina and Mecca became 

1 Tab. i. 1795. 


Muslims. With some of the others there may have been political 
alliances ; but some again had no connexion with Medina until after 
their defeat in the wars of the Riddah. Thus there was no apostasy, 
but at most political disloyalty. The supposed 'deputations' of all 
the tribes and their conversion are largely pious inventions to 
magnify the achievement of Muhammad (and perhaps to minimize 
that of Abu Bakr). 

This book is not the place for a full discussion of the Riddah, 
since it mostly falls in the caliphate of Abu Bakr. Nevertheless it 
has to be mentioned because it supplies information about the 
state of affairs during Muhammad's lifetime and because its 
beginnings are then. In addition to Muhammad's tribal policy, 
then, we are to give special consideration to two questions : firstly, 
to what extent the tribes were in alliance wifh Muhammad (even 
if the relationship was purely political); and secondly, how far 
the motives of the tribes were social and political, and how far 
religious. For this investigation we have, in addition to the his- 
torical narratives of Ibn Hisham and al-Waqidi, a collection of 
letters attributed to Muhammad and accounts of ' deputations' to 
him, preserved by Ibn Sa'd. 1 According to the critical principle on 
which this book is based, these are to be accepted as genuine except 
when they contradict other early source material or well-established 
facts. Careful attention, however, has to be paid to the precise 
assertions of the sources about the * deputations'. That a 'deputa- 
tion' came from a certain tribe does not mean that the whole of 
that tribe became Muslim, for often the 'deputation' would repre- 
sent nobody but themselves. An extreme case is that of the 'deputa- 
tion' from Ghassan. The story is presumably the best that could 
be found to maintain the honour of the tribe; but it is so poor 
a story that we may safely infer that no members of the tribe 
became Muslims during Muhammad's lifetime. 2 

The following are the main tribal groups mentioned in con- 
nexion with Muhammad, showing some of their genealogical 
connexions and how they have been divided for purposes of 

1 IS, i/2. 15-86; also edited and commented on by J. Wellhausen, Skizzen 
und Vorarbeiten, Berlin, 1889, iv/3 (whose paragraph number is added in brackets 
after the references to IS, i/2), and by Caetani; little is added by Sperber, 'Die 
Schreiben Muhammeds an die StSmme Arabiens', Mitteilungen des Seminars 
filr orientalischen Sprachen (Berlin), xix (1916), Westasiatische Studien, 1-93. 

2 Cf. p. 114 below. 


Tribes to the west of Medina and Mecca ( 2) 


Aslam; Ka'b b. *Amr; al-Mustaliq 

Bakr b. 'Abd Manat 
Damrah (with Ghifar); Layth; ad-Du'il; Mudlij 

al-IJarith b. c Abd Manat (part of Ahabish) 1 
Azd Shanu'ah (with Daws). 

Tribes to the east of Medina and Mecca ( 3) 

Khuzaymah (b. Mudrikah; Kinanah belonged to Khuzaymah) 

Asad b. Khuzaymah; ('Adal, al-Qarah) 
Tayyi' (with Nabhan) 
Hudhayl (b. Mudrikah) 


Muharib (b. Khasafah) 

Ashja c ; Fazarah; Murrah; Tha'labah (with Anmar, 'Uwal) 
Sulaym (with Ri'l, Shayban) 

'Amir b. Sa'sa f ah 

al-Bakka J ; Hilal; folab (with al-Qurta, 'Uraynah); Rabfah 

Jusham; Nasr; Sa'd b. Bakr; Thumalah 

Thaqif (B. Malik, Ahlaf) 


Tribes to the north ( 4) 

Sa e dHudhaym; f Udhrah 


Quda'ah (with Jarm, al-Qayn, Salaman) 



Lakhm (with Dar) 

1 The Ahabish were a collection of small tribes or clans, including at-Mu?taliq 
(of Khuzfi'ah) and al-Hdn b. Khuzaymah (with its subdivisions 'A^al and 
al-Qfirah); the leading group was al-H5rith b. 'Abd Manit b. KinSnah. They 
were closely attached to Quraysh. Cf. IH, 245; IS, i/i. 81. 8; M I Mecca, 154 ff.; 
pp. ao, 57 above and 83, 88 below. 

6788 G 



Tribes south of Mecca ( 5) 
Khath c am (and near it Azd Shanu'ah) 

'Ans; Ju'fi; Khawlan; an-Nakha'; Ruha'; Sa'd al-'Ashlrah 

(with Zubayd); Suda' 

al-tfarith b. Ka'b (with Nahd) 
Kindah (with Tujib) 

TT j \ perhaps not strictly tribal names 

Eladramawt j r r j 

'Akk and Ash f ar. 

Tribes in the rest of Arabia ( 6) 


Azd 'Uman; 'Abd al-Qays (in al-Bahrayn) 




Bakr (with Shayban) ; Taghlib. 


Muhammad's earliest supporters, apart from the Emigrants and 
the Ansar, were from the tribes roughly to the west and south-west 
of Medina. At his entry into Mecca in 630/8 he had in his army 
contingents from Sulaym, Ghifar, Aslam, Ka'b b. 'Amr, Muzay- 
nah, Juhaynah, Layth, Damrah, Sa'd b. Bakr, and Ashja'; 1 and 
of these all except the first and the last came from the district we are 
now considering; some at least of Sa'd b. Bakr seem to have been 
close to Layth, Damrah, and Ghifar in location, but it is genea- 
logically a part of Hawazin, and will be considered in 3, as will 
also Ashja' and Sulaym. Along with the remainder may be con- 
sidered al-Harith b. c Abd Manat (the central part of the Ahabish), 
Bakr b. 'Abd Manat with its subdivisions ad-Du'il and Mudlij, 
al-Mustaliq, and Azd Shanu'ah with its subdivision Daws. Several 

1 WW, 332. 


of these groups were related to one another. Ghifar was part of 
Damrah, while Damrah and Layth were parts of Bakr b. 'Abd 
Manat. The latter, again, along with al-Harith b. 'Abd Manat, 
belonged to Kinanah, of which Quraysh was reckoned a part. 
Another large group was Khuza'ah, which included Aslam, Ka'b 
b. c Amr, and al-Mustaliq. 

It is worth glancing at the past history of some of these tribes. 
Khuza'ah had at one time been masters of Mecca, but, together 
with their allies Bakr b. 'Abd Manat, had been attacked and 
expelled by Quraysh. 1 Khuza'ah fade into the background after 
this, but the hostility between Bakr and Quraysh continued until 
after the Hijrah. There was a war between them in which the 
Ahabish supported Bakr, though in the later war of the Fijar 
the Ahabish had gon^ over to the side of Quraysh. 2 The feeling 
between Bakr and Quraysh was stirred up again by the killing of 
the chief of Bakr to avenge a youth of Quraysh, and because of 
this Quraysh hesitated about going out to Badr until a prominent 
member of Bakr, Suraqah b. Ju'sham of Mudlij, said he would 
see that Bakr did not attack them in the rear. 3 At the same period 
Nawfal b. Mu'awiyah of ad-Du'il (part of Bakr) received a large 
sum of money from certain men of Quraysh to provide arms and 
camels, presumably for poorer members of his own tribe. Subse- 
quently this same Nawfal is frequently mentioned among the 
leaders of Quraysh. Thd closer relations between Quraysh and 
Bakr are perhaps mainly due to the threat from Muhammad. In 
630/8 Quraysh was ready to support Bakr when an old quarrel 
between Bakr and Khuza'ah flared up ; indeed Quraysh may have 
fanned the flames. 4 

All these subdivisions of Kinanah and Khuza'ah (along with 
Sa'd b. Bakr of Hawazin and Azd Shanu'ah) were within the sphere 
of the direct influence of Mecca. Juhaynah and Muzaynah were 
similarly within the sphere of influence of Medina. They seem 
to have been comparatively poor and weak, and incapable of inde- 
pendent action except on a small scale. They would not have dared 
to attack Medina as Ghatafan did. The only expeditions specifically 
directed against any of them were that against al-Mustaliq in 627/5 

1 Cf . Ml Mecca, 4 f . 

a Al-Azraqi, in Wustenfeld, Mekka, i. 71. 14; IS, i/i. 8z. 8-n; quoted in 
Ml Mecca, 155 f. 

3 IH, 430-3; WW, 43. 4 IH, 802 ff.; WW, 319-21 ; cf. p. 62 above. 


and one against a small section of Juhaynah in 629/8; the clan of 
Jadhlmah which Khalid was sent to 'summon to Islam 1 after the 
entry into Mecca is usually said to be Jadhlmah b. < Amir b. c Abd 
Manat b. Kinanah, but a Jadhlmah is also identified with al- 
Mutaliq of Khuza'ah. 1 Another mark of poverty might be the fact 
that members of Ghifar and Aslam looked after camels for Muham- 
mad, though this might also be due to the poverty of the individuals 
in question or to some other individual circumstances.* On the 
other hand some men of Ghifar seem to have had qualities of 
leadership something which is not common among the tribes 
considered in this section; Siba c b. 'Urfutah was left in charge of 
Medina during various absences of Muhammad. 3 

In the first year or two after the Hijrah Muhammad's chief aim 
must have been to gain friends, so that, whqn he and his followers 
went on expeditions, they could move about freely without fear of 
being molested. On an expedition in 623/2 he is said to have made 
a treaty with Mudlij and Damrah. 4 These two parts of Kinanah 
were presumably disaffected towards the Meccans, and therefore 
ready to undertake not to attack the Muslims. They may also have 
helped Muhammad by passing on information. The text of a treaty 
has been preserved, but it prescribes mutual help (nasr\ and that, 
unless it means less than it appears to mean, would be unlikely 
before Badr. 5 Another treaty with a section of Kinanah states that 
they are not to be required to help Muhammad against Quraysh. 6 

In his relations with Kinanah, Khuza'ah, and the other tribes 
in the neighbourhood of Mecca Muhammad was taking advantage 
of the conflicts of interest within the Meccan sphere of influence, 
in which Quraysh had been unable to effect reconciliation. The 
clearest case of this was after the treaty of al-Hudaybiyah when 
Khuza'ah allied themselves with Muhammad while their enemies, 
Bakr b. c Abd Manat, allied themselves with Quraysh. 7 In his 
approaches to these tribes Muhammad must have known or been 
able to discover which were dissatisfied with Quraysh. Until he 

1 For the usual view cf. Caetani, ii. 148; contrast IS, i/i. 46. 28. 

a Abu Dharr, Abu Ruhm, &c., of Ghifar WW, 227, 241 ; Salamah b. al- 
Akwa', Najiyah b. Jundub, &c., of AslamWW, 227 f. (== IH, 719*-), 241, 
300, 416. 

3 IH, 668 (= WW, 174); WW, 265; IH, 896 (- WW, 393), 966. 

4 IS, ul i. 5. 2, on the expedition of 'Ushayrah; cf. treaty with Pamrah alone 
during the expedition of al-Abw&' (WW, 34), which may or may not be the same. 

* Cf. Excursus F, no. i. 

6 Cf. ibid., no. 7. 7 Cf. p. 49 above. 


was .strong enough, however, to protect them from Qurayah, he 
could not expect them to join in any act of open hostility against 
Quraysh. Thus his first aim must simply have been to establish 
friendly relations. 

A second aim, however, gradually became evident, to gain 
supporters to join in his expeditions. To begin with he was more 
likely to find these in the Medinan sphere of influence. Juhaynah 
was confederate with the Khazraj at the battle of Bu'ath and 
Muzaynah with the Aws, 1 and the close relations continued. In this 
way Muhammad may be said to have had an indirect alliance with 
these tribes from his first days in Medina. In the first expedition 
from Medina the Muslims were helped by Juhaynah and are said 
to have been in alliance with them, but Juhaynah was also in 
alliance with Quraysh, and the help consisted in acting somehow 
or other so as to avoia a conflict. 2 There is no record of any impor- 
tant group within Juhaynah joining Muhammad as a group; the 
so-called 'deputation* (wafd) from Juhaynah seems to consist of 
two men speaking for themselves: 3 and the group with c Amr b. 
Murrah was probably small. 4 From an early period, however, 
individuals were attaching themselves to Muhammad. A reconnoi- 
tring party before Badr consisted of two men from Juhaynah, while 
a third was killed at Uhud. 5 There were also several men of Muzay- 
nah at Uhud, of whom one was killed. 6 Some of these, as con- 
federates of Medinan clans, may have lived most of the time in 
Medina. At a slightly later date there was a district of Medina 
inhabited by Juhaynah, and they had a mosque of their own. 7 It 
is probable, however, that many of those who became followers 
of Muhammad had hitherto been nomadic. A town-dweller, for 
example, would hardly have been put in charge of the pasturing 
of Muhammad's war steeds, a duty given to a man of Muzaynah. 8 
Very significant is a passage where a Juham who came to pledge 

1 A. P. Caussin de Perceval, Essai sur VHistoire des Arabes avant Vlslamisme, 
Paris, 1847-8, ii. 68 1 ff.; cf. attachments of individual members of Juhaynah: 
Basbas b. 'Amr was haltf of B. Sa'idah (Usd, i. 178 f.), 'Adi b. Abi Zughba' of 
B. Malik b. an-Najjar (ibid. iii. 394), Pamrah al-Juhani of B. Sa'idah (WW, 1 39), 
'Abdallah b. Unays of B. Salimah (Usd, iii. 1 19 f.). SinSn b. Wabr of B. Salim of 
'Awf (IS, iv/2. 70 = WW, 179). 

* IH, 419; WW, 33; cf. 44, where help is apparently given to Abu SufySn. 

3 IS, i/2. 67. 21 ff. 4 Ibid. 68. 3 ff. 

5 WW, 38, 44 Basbas b. 'Amr and 'Adi b. Abi Zughba'; WW, 139 Pamrah 
al-Juhani. * WW, 128, 139. 

7 IS, iv/2. 67. 12 f., 69. i ; ibid. i/2. 68. 2. 8 WW, 184. 


himself to Muhammad was asked whether he intended the be- 
douin pledge (bay^ ah 'arabiyah) or the pledge of Hijrah (bay 9 at 

A choice of this sort between continuing one's nomadic life and 
migrating to Medina (and taking part in expeditions) would be 
more critical for members of Kinanah and Khuza'ah than for 
members of Juhaynah and Muzaynah. The men from Ghifar and 
Aslam who looked after Muhammad's camels must have left their 
tribes. Moreover, such men would be direct adherents of Muham- 
mad himself, analogous to the Emigrants of Quraysh. Treaties are 
extant which speak of Aslam, Khuza'ah, and Muzaynah being 
classed as Emigrants, even when they did not leave their home 
districts; 2 and, though this probably belongs to the later years of 
Muhammad's life, it is an indication that individuals who had 
made the Hijrah or emigrated to Medina had the special status of 

Whatever may have been the position with such Emigrants from 
Juhaynah and Muzaynah, the first loyalty of those from the Meccan 
sphere of influence would be to Muhammad. Thus we find that 
the serious quarrel on the expedition to Muraysi' arose from a dis- 
pute between a Ghifari and a Juham; the Emigrants of Quraysh 
took the side of the Ghifari and the Ansar that of the Juhanl. 3 
Muhammad no doubt recognized that the accession of such recruits 
not merely strengthened the Muslim cause but also strengthened 
his own position within the Medinan community. The terms 
of agreement with a * mixed multitude' (jummc?) that was possibly 
little better than a robber band show Muhammad anxious to gain 
adherents. 4 By the time of his death there were probably many 
persons from Kinanah and Khuza'ah (and other tribes also) 
resident in Medina. 'Umar is said to have remarked that Abu 
Bakr would not have been acclaimed as caliph by the Ansar had 
not a group of Aslam come on the scene at the critical moment; 
this is probably an exaggeration the source of the information 
is a man of Khuza'ah but the tribesmen were a factor of the 
balance of power in Medina. 

As might have been expected, then, the tribes in the surround- 
ings of Medina and Mecca were among the foremost supporters 

1 IS, iv/2. 66. 3. 

IS, i/2. 24. 14-18 ( 29), 25. 1 1-27 ( 3*), 38 * ( 76) ; cf. 41 f. ( 79) J see also 
Excursus F, nos. 5, 8. 
3 IH, 626; WW, 179. 4 IS, i/2. 29. 13-22; cf. Excursus F, no. 6. 


of Muhammad after the Emigrants and the Ansar. They supplied 
contingents for his army at the conquest of Mecca and the battle 
of Hunayn. Only Aslam seems to have been present as a tribe at 
the earlier expedition of al-Hudaybiyah, though, despite the refusal 
of the tribes of Juhaynah, Muzaynah, and Bakr to participate, 1 
we find that many individuals from Juhaynah, for example, are 
said to have been present. 2 In later expeditions the same group 
of tribes were an important part of the Muslim forces. All remained 
faithful to Abu Bakr after Muhammad's death and showed no 
sign of defection. Along with the Emigrants and the Anar, 
therefore, they constituted the core of the Islamic state, and were 
fittingly granted the status and privileges of Emigrants. 


The tribes immecJiately east of Medina and Mecca may con- 
veniently be treated together as a second group, since after Quraysh 
they claimed most of Muhammad's attention in the years be- 
tween Badr and Hunayn. There is an important division among 
these tribes. Some were sufficiently friendly to the Meccans to be 
willing to join them for a consideration; Ashja', Fazarah, and 
Murrah (of Ghatafan) together with Asad b. Khuzaymah and 
Sulaym sent contingents to the siege of Medina. 3 Others again 
and notably Hawazin were hostile. More is known of the previous 
history of these tribes than of those in the last section, and a 
review of this contributes to the understanding of the position in 
Muhammad's time. 

Asad b. Khuzaymah was apparently closest to Quraysh. The 
eponymous ancestor is said to have been the sddin or priest of the 
Ka'bah at one time; 4 and some close connexion seems to be pre- 
supposed by the large number of confederates of 'Abd Shams (of 
Quraysh) from Asad b. Khuzaymah, 5 and by their alliance with 
Quraysh in the war of the Fijar. 6 The main part of the tribe, 
however, seems to have lived to the north and east of Ghatafan, 
where they were neighbours of Tayyi'. The latter had displaced 
them and constituted their main enemy. 7 There are several traces 

WW, 242. 

IS, iv/2. 66 ff. -Tamlm b. Rabi'ah, RSfi* b. Mukayth, Jundub b. Mukayth, 
Abu Dubays, Suwayd b. akhr. 

WW, 191. 4 Wustenfeld, Mekka, ii. 139 f. 

Cf. Ml Mecca, 174 f., where Khuzaymah is in fact Asad b. Khuzaymah. 
IS, i/i. 81. 8. 7 Cf. H. Reckendorf in El (i). 


of this feud during Muhammad's lifetime : it was a man of Tayyi' 
who brought the information in June 625 (i/4) that Tulayhah b. 
Khuwaylid and his brother were trying to raise Asad against the 
Muslims; it was a man of Asad who guided the expedition to 
destroy the god of Tayyi' ; and in a letter to Asad Muhammad 
warns them against trespassing on the waters and lands of Tayyi'. 1 
Similarly, in the wars following the death of Muhammad the 
Muslims found it easy to detach Tayyi' from Tulayhah. 2 

The main part of Asad was presumably at some considerable 
distance from Medina. Nevertheless the reports of Uhud en- 
couraged Tulayhah and his brother to collect a force to raid Medina 
before the Muslims recovered; Muhammad, however, forestalled 
him, and the Muslims by a lightning movement were able to 
capture some of the camels of Asad. This episode suggests that 
Tulayhah was an opportunist, and the result Aoubtless discouraged 
him from further thoughts of this kind while Muhammad lived. 
The only other expedition against Asad was some two years later, 
but was too small to be directed against Tulayhah; it may have 
been due largely to some feud within the tribe, since the Muslim 
leader was from that section of Asad which had become con- 
federates of *Abd Shams. Opportunism may also mark the sending 
of a deputation to Muhammad in the year 9 when there was a rush 
to 'get on the bandwaggon'. This would be especially the case if 
Tulayhah was a member of the deputation; 3 some two years later 
he was leader of Asad and other tribes in a war against the Muslims, 
and must already have had wide influence. It is noteworthy, how- 
ever, that there is a version which omits his name ; 4 if this is correct, 
it means that only a section of Asad submitted in 9 and that Tulay- 
hah remained in independence, doubtless making preparations to 
resist Muslim expansion; this seems most likely, but the question 
is complex. 

Little is heard of Khuzaymah apart from Asad. The small 
groups of 'Adal and al-Qarah belonged to al-Hun b. Khuzaymah 
and were part of the Ahabish, and so attached to Quraysh. 5 It was 
when they, allegedly in collusion with Sufyan al-Lihyanl, asked 

1 WW, 151; ibid. 389; IS, i/2. 23; cf. also WW, 152. 

a Caetani, ii/i. 131. The form Tulayfeah is not derisive; cf. p. 134, 
n. 3 below. * IS, i/2. 39. 

4 Usd t ii. 29, on Hatframi b. 'Amir, whose name comes first in the list in IS. 

5 MIMecca, 154 (A), 155 (K); cf. IS, ii/i. 39. n; WW, 155, 157, 199. There 
were men from al-Qrah in the/wmmJ', IS, i/2. 29. 15. 


Muhammad for men to instruct them in Islam that Sufyan was 
able to surprise the party. 

fayyi\ The main part of the tribe of Tayyi' (including the sub- 
tribe of Nabhan) was beyond Asad, and there were no expeditions 
from Medina against it apart from that led by 'All in July /August 
630 (iv/9) to destroy the god of the tribe, al-Fuls (or al-Fulus or 
al-Fils). Tayyi' is thus in a different position from the other tribes 
considered in this section, but it is convenient to mention it here 
in view of its feud with Asad. There were apparently some indi- 
vidual contacts between Tayyi' and Medina prior to Islam; the 
father of Ka'b b. al-Ashraf (an opponent of Muhammad, reckoned 
to the Jewish tribe of an-Nadir) belonged to B. Nabhan of Tayyi' ;* 
a man of Tayyi' is mentioned as acting along with the Medinan 
Hypocrites on the way back from Tabuk; 2 and two women of 
Tayyi' are recorcfed to have married into the Medinan clan of 
B. al-Harith. 3 An early convert from Tayyi' was Rafi' b. Abi Rafi' 
('Umayrah), who took part in the expedition of Dhat as-Salasil in 
October 629 (vi/8) as a Christian (or pagan), and accepted Islam 
soon afterwards. 4 In the two years following this many sections of 
the tribe seem to have professed Islam; letters have been preserved 
in which Muhammad guarantees their security provided they per- 
form the worship, pay the zakdt, hand over a fifth of any booty 
taken, obey his orders, and so on. 5 The most important convert 
was 'Adi, son of the well-known Hatim at-Ta'L When 'All's expedi- 
tion attacked the tribe, 'Adi made his way to Syria where, as a 
Christian, he expected to be well received. Muhammad, however, 
by means of his sister, persuaded him to come to Medina and 
become a Muslim; he was then put in charge of the sadaqdt of his 
tribe. 6 On Muhammad's death the tribe was apparently at first 
uncertain what attitude to adopt, but, when Asad was seen to be 
in the opposite camp, there was little difficulty in persuading them 
to support the Muslims. 

The behaviour of Tayyi' seems to be explained by the fact that 
it was largely Christian, but belonged to the Persian sphere of 
influence rather than the Byzantine. 'Adi may have been the readier 
to become a Muslim because he found the Byzantines uncongenial. 

1 Cf. p. 210 below. * WW, 408. 

3 IS, iii/2. 89, wife of 'Abdallah b. ar-Rabl'; viii. 264, mother of Kabshah 
bint Waqid. 4 IH, 985 ; WW, 315 f. 

5 IS, i/2. 23 ( 23), 30 ( 50, 59 f- ( 103); also 22 ( 21). 

6 IH, 947-50; IS, i/2. 60. 


The Christian Arab tribes of the east and north-east willingly 
entered into political alliance with Muhammad since both they and 
he were interested in raiding towards 'Iraq; and after the break- 
down of the Persian empire (in 628 and the following years) many 
became ready to accept Islam also. 1 Our information is so scanty, 
however, that it may well be that many members of Tayyi', though 
loyal to Medina during the Riddah, remained Christians till some 
time afterwards. 

HudhayL The tribe of Hudhayl had been involved in the affairs 
of Mecca in pre- Islamic times. Their chief is said to have accom- 
panied e Abd al-Muttalib when he went to negotiate with Abrahah. 2 
They had blood-feuds with Layth (Kinanah) and Aslam and Ka'b 
(Khuza'ah), 3 but these may have been temporary, for it was 
possible about the year 626/4 for a Medinan Muslim to win the 
confidence of the chief Sufyan by pretending to belong to Khu- 
za'ah. 4 This chief, Sufyan (or Ibn Sufyan) b. Khalid b. Nubayh 
of the sub-tribe of Lihyan, was collecting a force to raid Medina, 
but 'Abdallah b. Unays was sent by Muhammad and assassinated 
him. It was apparently in revenge for this that B. Lihyan attacked 
the Muslim instructors asked for by their friends the clans of 
'Adal and al-Qarah. Some were killed, but the prisoners were sold 
to Quraysh to sate their desire for revenge, an indication of the 
close ties between Hudhayl and Quraysh. 5 In view of this history 
it is not surprising that some of them offered active resistance at 
the conquest of Mecca. 6 They were probably incorporated in the 
Islamic state on similar terms to Quraysh, but the sources mention 
only their presence in Mecca and at the siege of at-Ta'if, and the 
destruction of their idol by 'Amr b. al-'As. 7 

Muharib. The small tribe of Muharib is obscure. It was attacked 
in the course of three expeditions in the years 624/3 an d 627/6, and 
it sent a deputation to Muhammad in 63i/io. 8 It is presumably 
Muharib b. Khasafah (though there is also a minor clan of Quraysh 
called Muharib), but it seems to have become linked with Ghat- 
afan, and in particular with Tha'labah; part of it is said to have 
lived among Tha'labah. 9 It is coupled with the latter in the accounts 

1 Cf. p. 131 below. IH, 34. 

3 WW, 369 and IS, iv/i. 32; WW, 341 f. 4 Ibid. 225. 

IH, 981 f. = WW, 224 f. and Caetani, i. 577 f- ; IH, 638-48 = WW, 156-60. 

6 WW, 333 f- 7 Ibid. 342, 369, 35<>. 

8 Ibid. 99 *., 2*6, 233; IS, i/2. 43 ( 83). 

9 Aghdnl, xii. 124. 26; cf. Wellhausen's note on 83. 


of two of the Muslim expeditions, and the leader Du'thur b. al- 
Ilarith is sometimes called Ghatafani. 1 Although genealogically 
distinct it seems to have become in practice a part of Ghatafan, 
and need not be further considered separately. 

Ghatafan. In Muhammad's time Ghatafan was a collection of 
tribes rather than a single tribe. To it belonged 'Abs and Dhubyan 
between whom the celebrated war of Dahis had been fought. 'Abs 
plays little part in the events of Muhammad's lifetime, 2 but much 
is heard of the subdivisions of Dhubyan, namely, Fazarah, Murrah, 
and Tha'labah. 3 Ashja' formed a distinct branch. 

On the conclusion of the war of Dahis there was a genuine 
reconciliation between e Abs and Dhubyan, and we hear of little 
further strife within Ghatafan. Instead there is a series of wars 
with Hawazin (Jusham, Nasr, and 'Amir) or rather with the wider 
group of Khasarah which also included Sulaym. Sometimes 
most of Ghatafan and Khasafah were involved, sometimes only 
one or two tribes on each side. We hear of fighting between 
Sulaym and Murrah, between Fazarah and 'Amir, between 'Abs 
and 'Amir, between Fazarah (with c Abs) and Jusham, and so on. 
In some of these 'days' Fazarah was led by 'Uyaynah b. Hisn, 
the great-grandson of that Hudhayfah b. Badr who had success- 
fully led Dhubyan in the war of Dahis; thus he was the heir 
of a tradition of leadership. 4 A large part of the relations of 
Muhammad with Ghatafan are those with 'Uyaynah. Since 
Hawazin had been the chief enemy of Ghatafan for some decades 
and was also an enemy of Quraysh, we might expect a drawing 
closer of Ghatafan and Quraysh. The hostility of Ghatafan and 
Sulaym had not been so bitter, and the first Muslim expedition 
against these tribes is said to have been directed against a mixed 
group of the two; nevertheless there was rivalry between the 

In September 624 (iii/3), about two months after the first expedi- 
tion just mentioned, Muhammad himself led a large force against 
Tha'labah and Muharib. Though there was no full-scale encounter 
with the enemy, the expedition seems to have had a deterrent effect, 
since no further trouble with Ghatafan is recorded until nearly two 
years later when they brought about 2,000 men to join in besieging 

1 Usd, ii, s.v. * Cf. IS, i/2. 41, a convert from 'Abs. 

3 With Tha'labah are mentioned Anmar and 'Uwal, otherwise obscure. 

4 Caussin de Perceval, ii. 424, 536-68, &c. 


Medina (April 627 = xi/5). Mention is even made of a temporary 
truce with 'Uyaynah. 1 While it was doubtless the apparent strength 
of Muhammad that induced 'Uyaynah to make such a truce, the 
prospect of a grand alliance against Muhammad, coupled with 
the diplomatic pressure and bribes of the Jews from Khaybar (who 
included the exiles of an-Na4ir from Medina), attracted him to 
active opposition. When Muhammad's trench prevented the alli- 
ance from making full use of their numbers in attack and forced 
them to engage in a siege, 'Uyaynah was tempted to consider 
Muhammad's offer of a third of the date-harvest of Medina in 
return for his immediate withdrawal. What exactly happened is 
not clear, 2 except that the parties in the alliance became more 
suspicious of one another and that 'Uyaynah did not get his dates 
from Medina. It was perhaps because he felt he had been unfairly 
treated in this matter that in August of the same year (iv/6) he 
raided Medina, causing great perturbation but getting the worst 
of the exchanges. 

Meanwhile one of the results of the failure of the besiegers was 
that Ashja' went over to Muhammad. Nu'aym b. Mas'ud of Ashja' 
had become a Muslim about the beginning of the siege, but had 
not made the fact public, and so, as a secret agent for Muhammad, 
had done much to increase the dissension between Quraysh, 
Ghatafan, and Qurayzah (there may, however, be some exaggera- 
tion as the story comes from Nu'aym^ himself through Ashja'I 
transmitters). 3 The deputation which came to Muhammad after 
the siege is said to have been headed by Mas'ud b. Rukhaylah, the 
leader of the contingent from Ashja' which had taken part in the 
siege. 4 The reasons they gave for coming to Muhammad sound 
genuine ; they were the section of their tribe (Ghatafan ? or Ashja' ?) 
living closest to Muhammad and smallest in number, and they 
were in distress through the war. They might have added that they 
had been allies of the Khazraj at Bu'ath. 5 It is doubtful, however, 
whether the whole tribe submitted to Muhammad at this time. 
A secondary report in Ibn Sa'd gives the figure of 700 men, but the 

1 WW, 182 foot. 2 Cf. p. 38 above. 3 IH, 680-2; WW, 205 f. 

4 IS, i/2. 48. 26 f.; cf. iv/2. 19-24, biographies of ten early converts from 

5 Caussin de Perceval, ii. 68 1 ; cf. 657, a Tha'labi as a client of Malik b. al- 
'AjlSn. WW, 233 mentions a Ghatafani on the small expedition led by Muhammad 
b. Maslamah in August 627 (iv/6), perhaps an old confederate of the latter's; 
Caetani (i. 694 n.) suggests he was not a Muslim. 


main one gives only a hundred, while the contingent at the siege 
was said to number 400. Moreover there is some confusion of names. 
Nu c aym b. Mas'ud is sometimes called Nu'aym b. Mas'ud b. 
Rukhaylah, which would make him the son of the chief; 1 but it 
is more likely that the true version is that which gives his grand- 
father as 'Amir. z There is no further mention of Mas'ud b. Rukhay- 
lah, but Nu'aym b. Mas'ud appears frequently as one of the two 
leaders of Ashja' among the Muslims. It seems probable, therefore, 
that he should be given as the leader of the 'deputation', especially 
as he is named in the text of an agreement with Muhammad which 
looks as if it refers to this occasion. 3 There are said to have been 
300 men from AshjY with Muhammad at the conquest of Mecca, 
and there is no word of Ashja' joining in the war against Abu Bakr. 
After 'UyaynahJs mid on Medina in August 627 we hear nothing 
of him until the Khaybar campaign in May and June 628 (i/7). 4 
In the interval there had been three small expeditions against 
Tha'labah and their neighbours, and one led by Zayd b. Harithah 
against a section of Fazarah to take revenge for their ambushing 
him and his party. 'Uyaynah remained on friendly terms with the 
Jews at Khaybar. At least when they were attacked by Muhammad 
he was prepared to bring 4,000 men to support them in return 
for half the date-harvest of Khaybar. His support, however, was 
not whole-hearted and consequently ineffective. What exactly 
happened is not clear, but 'Uyaynah certainly negotiated with Mu- 
hammad. One account is that Ghatafan retired because they heard 
that the Muslims were attacking their families in their rear ; this was 
perhaps a rumour spread by Muhammad. Another account is that 
he offered them an equal or larger quantity of dates; and as the 
Jewish strongholds began to fall this would be more valuable than 
the similar Jewish promise. Yet another story says that 'Uyaynah 
received a part of Khaybar called Dhu 'r-Ruqaybah from Muham- 
mad. The three stories are not irreconcilable; it may be that 
Muhammad refused to keep his offer of dates because 'Uyaynah 
had withdrawn (or Muhammad alleged he had withdrawn) to meet 
a possible attack on the families in the rear, and that he gave him 
the piece of land as compensation. 

1 WW, 205; IS, i/2. 26. 19. 

2 IH, 680; Usd and Ifdbah, s.v.; IS, iv/2. 19-21. 

3 IS, i/2. 26. 18-20; cf. Excursus F, no. 10. 
* IH, 755-81 ;WW, 264-96. 


If this was how 'Uyaynah was treated, it would explain why in 
February 629 (x/7) he collected some men 'with hostile intentions'. 
These cannot have amounted to more than a raid, however, for 
only 300 men were sent against him. Perhaps the growing military 
reputation of the Muslims and his own lack of success were making 
the tribesmen reluctant to follow him. Al-Harith b. 'Awf, the leader 
of Murrah, who had been responsible for reconciling 'Abs and 
Dhubyan, is said on this occasion to have counselled him to submit 
to Muhammad; at Khaybar also he seems to have been unwilling 
to fight. 1 Moreover in December 628 and January 629 (viii, ix/y) 
sections of Murrah and Tha'labah had been roughly handled by 
Muslim expeditions. Apart from a very small expedition against 
Jusham in December 629 (viii/8) of which no further details are 
recorded, there was no further fighting between Ghatafan and the 
Muslims until after Muhammad's death. They had learnt that 
there was no profit to be made by opposing Muhammad. 

It would seem that there must have been some agreement about 
this time between Muhammad and 'Uyaynah, although there is no 
mention of any in the sources. 2 'Uyaynah was with the Muslim 
army at the conquest of Mecca and the battle of Hunayn in January 
630. Although he apparently did not have even a small detachment 
of his tribe with him, 3 he was treated honourably, was able to use 
his influence on behalf of a man of Ashja', 4 and received a hundred 
camels at al-Ji'ranah. In April or May 630 (i/9) Muhammad 
accepted his offer to punish a small part of Tamim which had refused 
to pay their dues to Muhammad's agent, and this is reckoned one 
of the Muslim expeditions. 5 Such agreement as 'Uyaynah may 
have made, however, need not have involved becoming a Muslim. 
There is no record of his having accepted Islam. About January 
631 (x/9), after the expedition of Tabuk, there were 'deputations' to 
Medina from Murrah and Fazarah. 6 The former was led by the 
chief al-Harith b. 'Awf, and the latter by 'Uyaynah's brother and 
nephew, Kharijah b. Hisn and al-Hurr b. Qays b. Hisn. The 
absence of 'Uyaynah is noteworthy and together with his conduct 
in 630 presupposes an earlier agreement between him and Muham- 
mad. It may well be, as Caetani suggests, that this earlier agreement 
is passed over in silence because 'Uyaynah was allowed to remain 

1 WW, 299, 270. 3 Cf. Caetani, ii. 447. 

* WW, 327. 4 Ibid. 366. 5 IH, 933-8; WW, 385-7. 

6 18, i/2. 42; Tha'labah had accepted Islam about March 630, ibid. 43. 


a heathen. When after an inglorious share in the Riddah he was 
captured and taunted by the boys of Medina with going back on 
his faith, he is said to have remarked, *I never believed in God'. 1 
There was probably no serious rift between 'Uyaynah and his 
brother Kharijah, and both took part in the Riddah. Yet 'Uyaynah 
may at times have acted separately in hopes of gaining some private 
advantage, and Kharijah seems to have been the independent com- 
mander of a section of the tribe. 2 

There is an amusing story which, whether literally true or not, 
is probably an excellent characterization of 'Uyaynah. At al- 
Ji'ranah he selected as his prize an old woman for whom a large 
ransom might be expected. Her son offered 100 camels for her but 
'Uyaynah wanted more. Later, however, 'Uyaynah thought better 
of it and said he >would accept 100, only to be presented with 
a reduced offer of 50, which he refused. Again he thought better 
of it, but the offer was now reduced to 25. The same thing happened 
again, and the offer sank to 10. By the time 'Uyaynah was ready to 
accept this, the woman's son was asking for her to be set free without 
ransom. Eventually, though with a bad grace, 'Uyaynah did so, 
only to be met with the request for the present of a dress ; and in 
the end a dress he had to give. 3 Greed, which led him to try to drive 
too hard a bargain, coupled with lack of judgement, which often 
made him fare badly, are to be traced in all his relations with 
Muhammad. It is not surprising that in the pages of the slrah and 
the history of the Riddah, especially in al-Waqidi, he is presented 
as something of a laughing-stock. By combining severity with 
kindness on appropriate occasions, Muhammad was able to detach 
'Uyaynah from Quraysh, but he did not manage to incorporate his 
tribe securely in the Islamic state. 

Sulaym. The tribe of Sulaym, as has been noticed, sometimes 
joined with Hawazin against Ghatafan. It was perhaps because of 
the memory of this alliance that Sulaym showed so little spirit 
against Hawazin at Hunayn (if the report is unbiased). 4 Latterly, 
however, they had come to be closely linked with Quraysh, espe- 
cially 'Abd Shams and Hashim. 5 This was due above all to the 

1 Tab. i. 1897; further references in Caetani, ii/i. 622. 
a Ibid. ii/i. 592 f.; I$dbah, s.v. Kharijah; cf. W. Hoenerbach, Watima's Kitdb 
ar-Ridda aus Ibn I?agar*s I$dba, Wiesbaden, 1951, 83 f. 3 WW, 378 f. 

4 Ibid. 358, &c. ; but a Sulami killed Durayd b. as-Simmah of Jusham (ibid. 


5 H. Lammens, La Mecque d la Veille de VHtgire, Beirut, 1924, 196-8; 


fact that there were gold-mines in the territory of Sulaym and that 
Quraysh were able to help in their development. 1 Shortly before 
Sulaym became involved with the Muslims the leadership had been 
disputed by Khufaf b. Nadbah and al-'Abbas b. Mirdas, and by the 
award of an arbiter from Hawazin the functions had been divided 
between them. 

There were Muslim expeditions against Sulaym in July and 
October/November 624 (i, v/3), of 200 and 300 men respectively. 
Some Ghatafan are also said to have been involved in the first. 
On the second the enemy dispersed before contact was made. After 
this there does not seem to have been any widespread hostility to 
Muhammad among Sulaym. Two clans of Sulaym were certainly 
responsible for the massacre of Bi'r Ma'unah in July 625 (ii/4), and 
700 men of Sulaym are said to have joined in besieging Medina as 
a result of Jewish bribes and intrigues. The other two expeditions 
mentioned in the sources cannot have been against Sulaym as a 
whole. One was a little-known expedition about September 627 
(iv/6) led by Zayd b. Harithah. 2 The other, led by a man of Sulaym 
in April 629 (xii/7), was probably a dispute within the tribe which 
was taken under the aegis of Islam. 3 For the conquest of Mecca and 
the battle of Hunayn in January 630 (ix/8) Sulaym provided 900 
or 1,000 men. All this tends to show that there was a party favour- 
able to Muhammad in Sulaym, so that it was never necessary for 
him after 624 to keep the tribe quiet by a,show of force. Among the 
Emigrants at Medina were a few confederates from Mecca belong- 
ing to Sulaym, and this may have helped. On Muhammad's death 
only a few clans of Sulaym rebelled; the main part of the tribe 
remained loyal. 4 

How Sulaym came over to Islam is obscure, and also the precise 
part played by al-'Abbas b. Mirdas. In one story that occurs 
among the accounts of the 'deputations' credit for the conversion 
of the tribe is claimed by a member of the leading clan of ash- 
Sharid, Qidr b. 'Ammar. 5 (This is conceivably a mistaken form of 
the name of the well-known chief of the tribe, Sakhr b. 'Amr, but 

cf. Ml Mecca, 175 ; the wife of 'Uthman b. Maz'un was of Sulaym, but her mother 
was of 'Abd Shams (IS, iii/i. 286; WW, 372). 

1 The sanctuary of al-'UzzS at Nakhlah belonged to B. Shayban of Sulaym, 
WW, 351, &c. 

a Tab. 1555; cf. Caetani, i. 694 f. Not in IH and WW. 

3 WW, 303 ; not in IH. 

4 Cf. Caetani, ii/i. 579 f. IS, i/2. 50. 4. 


he is thought to have died before the Hijrah.) 1 Qidr was alleged 
to have promised to help Muhammad with 1,000 men. On his 
death he charged three men, al-'Abbas b. Mirdas, Jabbar b. al- 
Hakam, and al-Akhnas b. Yazld 2 to carry out his promise, and this 
they did. In other accounts, however, Sulaym was summoned to 
join the expedition to Mecca by al-Hajjaj b. 'Hat and al-'Irbad b. 
Sariyah, while the three standards at Hunayn were carried by 
al-'Abbas b. Mirdas, Khufaf b. Nadbah, and al-Hajjaj b. 'Hat. 3 
It is also significant that on several occasions men of Sulaym 
appear under the leadership of Khalid b. al-Walid and possibly 
suffer from the later tendency to blacken his name. 4 It seems 
certain, then, that there was no undisputed leader of Sulaym, and 
that for this reason Muhammad originally gave al-'Abbas b. Mirdas 
an ordinary man's sbare of four camels, instead of the hundred 
camels given to chiefs. 5 This situation may have made it easier for 
Muhammad to win over Sulaym, and his dealings with this tribe 
were certainly one of his successes. 6 

'Amir b. Sasaah. The tribe of 'Amir b. Sa'sa'ah (not to be 
confused with B. 'Amir of Quraysh) was a part of Hawazin, but is 
sufficiently important to be treated separately. In pre- Islamic times 
it had fought against various parts of Ghatafan and also against 
Tamim. An incident between it and Quraysh (with Kinanah) is also 
recorded. 7 The chief for many years had been Abu '1-Bara' 'Amir 
b. Malik, but latterly, though he kept the name of chief, some of 
the power was in the hands of two younger men, 'Amir b. Tufayl 
and 'Alqamah b. 'Ulathah. 8 Some of the consequences of this 
divided rule are to be seen in the affair of Bi'r Ma'unah; Abu 
J l-Bara' had given a safe-conduct to the Muslim party, but this did 
not prevent 'Amir b. Tufayl from instigating two clans of Sulaym 
to attack them. 9 

There were practically no attacks on B. 'Amir by the Muslims. 
In June 627 (i/6) a small party of thirty men captured some booty 

1 Caussin de Perceval, ii. 556-63, &c. 

2 Presumably al-Akhnas b. Hablb. Cf. IS, iv/2. 17. 1 6 and Usd, iv. 402 top 
(s.v. Ma'n b. Yazld); but in Usd y i. 56 al-Akhnas b. Khabbab. 

3 WW, 326, 358; for al-Hajjaj cf. ibid. 289. 4 Ibid. 351 f-, 358, 363. 

5 IH, 881 f.; WW, 376. Other references to al- f Abbas: IH, 832 (Caetani, ii/i. 
i 4 7f.);WW,2 7 9,378. 

6 Early converts from Sulaym, IS, iv/i. 157-60, iv/2. 14-19; cf. Safwan b. al- 
Mu'aftal, involved in the 'affair of the lie', IH, 732, WW, 185, &c. Letters about 
land. rights: IS, i/2. 26, 34 ( 34, 64, 65). 7 Caussin de Perceval, i. 298 f. 

8 Ibid. ii. 564-8. 9 Cf. p. 31 above. 

6783 H 


from a section of Kilab, and about three years later there was a raid 
on the same section under the leadership of ad-Dahhak b. Sufyan, 
himself of Kilab. The small expedition to Sly in July 629 (iii/8) 
was presumably against members of 'Amir, since Sly was in the 
territory of 'Amir, but the sources say merely that the people 
attacked belonged to Hawazin. That is all that is recorded. It may 
be that their high reputation as cavalry deterred the Muslims. 1 It 
is more probable, however, that Muslim policy was due rather to 
the readiness of at least a section of the tribe of f Amir to be friendly 
with Muhammad; this attitude may be partly the result of their 
hostility to the Meccans or of a friendship with the Medinans which 
is perhaps to be inferred from their alliance with an-Nadir. 2 What- 
ever the reasons on each side, there had been some agreement 
between Muhammad and Abu '1-Bara' ; at the l^ast this meant that 
each was to grant protection (jiwdr) to the followers of the other. 
There is no word of B. 'Amir having joined in the siege of Medina, 
and, thanks to the son of Abu '1-Bara' there were only a few (and 
these of Hilal) with Hawazin at the battle of Hunayn. 3 

After Bi'r Ma'unah 'Amir b. Tufayl submitted to Abu '1- Bara' ; 
a son of the latter wounded him because he had disregarded the 
grant of protection, but he claimed no revenge for the wound, and 
so closed the incident; Muhammad's payment of blood-money to 
him no doubt helped. He is also said to have asked Muhammad, 
as a reward for becoming a Muslim, to grant him the succession to 
Muhammad's position ; when this was refused he turned away, and 
probably died soon afterwards, since nothing further is heard of 
him. 4 His rival 'Alqamah b. 'Ulathah, also went to Muhammad; 
the report that he and his companions took the oath of allegiance 
on behalf of 'Ikrimah b. Khasafah (that is, a group of tribes in- 
cluding Hawazin and Sulaym) perhaps indicates that they were 
hoping that Muhammad would establish them as leaders of this 
group of tribes. 5 'Alqamah was evidently disappointed in Muham- 
mad, for after the siege of at-Ta'if in February 630 (x-xi/8) he is 
said to have gone to Syria and only to have returned after the 
death of Muhammad. He and Qurrah b. Hubayrah seem to have 
been the only men of note in 'Amir b. Sa'sa'ah who opposed Abu 
Bakr in the Riddah. When the military superiority of the Muslims 
became clear, they quickly surrendered or were captured, and were 

1 IS, i/2. 51. 22. 3 IH, 652; WW, 160. ' WW, 355. 

4 IS, i/2. 51. 5 Ibid. 52. 25; WW, 306. 


pardoned by Abu Bakr; some of their followers were punished for 
war crimes according to the lex talioms. 1 The rest of the tribe 
presumably the major part is said to have avoided taking a 
definite attitude until the situation cleared. This probably is a mark 
of caution rather than of active disaffection to Islam. 

On the whole, then, Muhammad was successful in winning 
'Amir b. Sa'sa'ah to his side. They do not seem to have become 
Muslims, however, until comparatively late, and then only a few. 
Ad-Dahhak b. Sufyan is spoken of as having summoned Kilab to 
Islam. His distribution of the alms (sadaqah) among the poor of 
his own tribe may be an indication that B. 'Amir regarded them- 
selves as equal allies of Muhammad and not as subject to him. 2 

lHawdzin. Hawazin properly includes 'Amir b. Sa'sa'ah and 
Thaqif, which are hey e being considered separately. The Arabic 
sources, however, sometimes seem to use the name when they are 
referring to the smaller group correctly known as 'Ujz Hawazin, 
'the rear of Hawazin', which omitted 'Amir and possibly Thaqif, 
and comprised the sub-tribes of Nasr, Jusham, Sa'd b. Bakr, and 
the smaller Thumalah. 3 Political alignments did not exactly follow 

1 Caetani, ii/i. 577, 603 f., 619-22. 

2 IS, i/2. 44. 16; 'deputations' from parts of 'Amir, ibid. 44-47, 50-52. Ao!- 
Pahhak b. Sufyan b. e Awf al-Kilabi (IS, iv/i. 29. 5 and Usd) is usually distin- 
guished from ad-pahhak b. Sufyan b. al-Harith as-Sulami (IS, iv/2. 17 and 
Usd). But it is curious that the former has many connexions with Sulaym, while 
the second is said to have had a standard (favd') at the conquest of Mecca, 
although he is not mentioned among the standard-bearers of Sulaym. Perhaps 
the Kilabi had a standard, and, because there were no Kilab present, some early 
scholar tried to explain the statements by inventing a Sulami; or perhaps the 
Kilabi belonged to Sulaym on his mother's side. 

Living with Kilab was the small tribe of 'Uraynah. Some stole Muhammad's 
camels about March 628 (x/6) and were punished. Later 'Abdallah b. 'Awsajah 
al-'Uram was used by Muhammad as a messenger, WW, 388; but cf. IS, i/2. 
3i (52). 

3 Caetani, ii/i. 57; O. Loth in Zeitschrift der Deutschen Morgenldndischen 
Gesellschaft, xxxv. 596; Buhl, Muhammed, 298 n. The tribe of Sa'd b. Bakr had 
apparently two sections. Some were present with Muhammad at the conquest 
of Mecca, having been summoned through Ghifar (WW, 326, 332); at the battle 
of Hunayn there were 200 along with Pamrah and Layth. Others were with 
Hawazin at Hunayn (IH, 840; WW, 364). The leader of the deputation to 
Muhammad, Zuhayr b. urad (Abu urad or Abu Jarwal or Abu Tharwan) is 
curiously called al-Jushami as-Sa'di in Usd, ii. 208, though Sa'd was not genea- 
logically part of Jusham. Perhaps some had become attached to Pamrah and 
some to Jusham. If al-W3qidi is right in dating the 'deputation* of Pimam b. 
Tha'labah (IS, i/2. 44) in vii/5 (December 626), this would presumably be from 
the section attached to Pamrah; but the date is unduly early, and the story has 
the marks of later editing (cf. Caetani, i. 609 f.). 


genealogy, and parts of 'Amir such as Hilal were sometimes found 
with 'Ujz Hawazin though the rest of 'Amir stood apart. In earlier 
times Hawazin's great enemy had been Ghatafan. More recently 
it had fought two bitter wars with Quraysh and Kinanah, the wars 
of the Fijar, and the memory of these was doubtless still a factor 
of importance. 

i Until the campaign of Hunayn in January 630 (x/8), Muhammad 
ad had almost no contacts with Hawazin, so far as our records go. 
Apart from those with 'Amir mentioned above there had been 
only two expeditions about December 628 (viii/y), led respectively 
by Abu Bakr and 'Urnar. These expeditions and the extension of 
Muhammad's sway over Sulaym (manifested by the presence of 
a contingent at the conquest of Mecca) may have made Hawazin 
aware that the growing Muslim strength was a, threat to them. On 
the other hand, when they heard that Muhammad was marching 
on Mecca, they may have expected a bloody and indecisive battle 
and may have hoped to have an easy victory over one or both of 
the exhausted combatants. Perhaps the two motives were com- 
bined. Muhammad cannot have planned to attack them until 
they themselves massed their forces. With an unwieldy army 
such as he was then leading he could have accomplished little 
against a mobile and widely dispersed foe. Once they presented 
him with a target, however, he would gladly seize this oppor- 
tunity of providing the booty which his army had not received in 

The battle of Hunayn, which has already been described, re- 
sulted in the defeat of Hawazin and the capture of their families 
and animals, though their allies of Thaqif managed to retreat safely 
to at-Ta'if. Muhammad entered into negotiations with Hawazin, 
and eventually came to an agreement with them. The negotiations 
were apparently carried out to begin with by men of Sa f d b. Bakr, 
who had milk-relationship with Muhammad since his wet-nurse 
Halimah had belonged to that tribe. 1 Through these men Muham- 
mad conveyed to the leader of Hawazin, Malik b. 'Awf an-Nasri, 
the generous terms on which he was prepared to make peace. Malik 
thereupon escaped from at-Ta'if and came to Muhammad to 
express his acceptance. He was to receive back his family and 
property, and to be given a present of a hundred camels; he was 
also to be recognized as chief of those sections of his tribe which 

' IH,8 7 7;WW, 377;IS,i/i.?2. 


had become Muslim. 1 The only obligation upon him that is expli- 
citly stated is that he was himself to accept Islam. There is mention 
elsewhere, however, of a payment called si'dyah made by Nasr, 
Sa'd b. Bakr, Thumalah, and Hudhayl Nasr being Malik's own 
tribe, and Thumalah one of the tribes or clans over which he was 
recognized as chief; since si* ay ah is used for the work done by 
a slave to earn his emancipation, the payment was probably not 
strictly a tax, but a payment for the liberation of the women. 2 
Malik in fact also attacked Thaqif, but this may have been not as 
part of the agreement but from inclination, because he was annoyed 
with them for deserting him, and in order to make up for his 
tribe's loss of camels. 

In the wars of the Riddah no part was taken by Hawazin. They 
are said to have been \mcertain for a time, and to have suspended 
paying the legal alms (sadaqah), but there is no mention of any of 
them taking up arms. 3 Malik was no doubt genuinely reconciled 
to Muhammad and the rule of Medina. Besides that, however, 
there was an important political reason. Hawazin must have been 
comparatively weak after the loss of their property at Hunayn. 
The rebels nearest to them were Asad and Ghatafan, and to sup- 
port these, their ancient rivals, would mean substituting their yoke 
for that of Quraysh and the Muslims. Self-interest thus helped to 
restrain Hawazin from armed revolt. But self-interest would not 
have led them to decide in this way, since Quraysh recently had 
been bitter enemies, had not Muhammad's careful handling of 
Malik after Hunayn shown him and his tribe that they were likely 
to be better off within the Islamic community than as members of 
a confederacy headed by Tulayhah of Asad. Muhammad's conduct 
had made it clear that, even when he was fighting Malik, he was 
hoping to win him over to support him, for he had arranged for 
Malik's family not to be distributed as booty but to be kept together 
in safety in the house of a kinswoman in Mecca. 4 The reluctance 
of Hawazin to combine with other tribes against Abu Bakr is a mark 
of the success of Muhammad's policy towards them. 

Thaqif. The tribe of Thaqif which inhabited the city of at-Ta'if 
consisted of two sections known an Banu Malik and the Ahlaf . The 
name of the latter group means * confederates' or 'confederacies', 

' IH, 879 ;ww, 379. 

2 IS, i/2. 24. 5-14 ( 28); for si ayah cf. Lane s.v. sa'd, i, 10. 

3 Tab. 1871. 4 WW, 379. 


but does not indicate an inferior social or political status. The 
Ahlaf may have been less influential in politics than the Band 
Malik, but that is a different matter. Their weakness they made up 
for by friendship with Quraysh. B. Malik, on the other hand, were 
associated with Hawazin. In the early years of Muhammad (as has 
been noted more than once) the wars of the Fijar between Hawazin 
and Thaqif on the one hand and Quraysh and Kinanah on the 
other hand had resulted in the control by Quraysh of the commerce 
of at-Ta'if. There was doubtless resentment among Thaqif at this 
state of affairs, and Muhammad may have hoped to profit by this 
when he visited them before the Hijrah. 1 

Whatever the past history of these two groups may have been, 
there is no doubt about their attitude towards Quraysh. When we 
hear that Thaqif sent a hundred men to hlp Quraysh at Uhud, 
we may safely assume that these were mainly from the Ahlaf. 2 
There were presumably some Thaqif at the siege of Medina, though 
there is no record of this. The employment of 'Urwah b. Mas'ud 
(of the Ahlaf) in the early negotiations at al-Hudaybiyah was made 
possible by the presence of a detachment of Thaqif with Quraysh 
there. 3 'Urwah was presumably aware of the repercussions that the 
decline of Meccan power would have on the politics of at-Ta'if and 
had been working to avoid a break with B. Malik at this juncture. 
When al-Mughirah b. Shu'bah of the Ahlaf had killed thirteen of 
B. Malik and then become a Muslim, that is, taken refuge with 
Muhammad, 'Urwah had reached an understanding with Mas'ud 
b. 'Amr of B. Malik to avoid bloodshed and had himself accepted 
responsibility for paying the blood-wit of 1,300 camels. 4 It was 
presumably also the growing weakness of Quraysh that forced the 
Ahlaf to acquiesce in the decision of the B. Malik to join Hawazin 
against Quraysh in January 630. Both parties had contingents at 
Hunayn, but it is not surprising to learn that, while B. Malik fought 
stubbornly and lost a hundred men, the Ahlaf took to flight when 
they saw the Muslims rallying and standing firm, and had only two 

During the battle of Hunayn and the subsequent siege of at- 
Ta'if 'Urwah had been in a city on the Byzantine frontier learning 

1 H. Lammens, La Cite" Arabe de Tdif d la Veille de VHegire, Beirut, 1922 
(from Melanges de la Facultt Orientate de Beyreuth, viii), iO4/2i6ff., but not 
to be accepted without reserves; cf. MJ Mecca, 139. 

2 WW, 102. ' IH , 744. 2 ; cf. WW, 250-2. 4 Ibid. 


about siege-engines and protective measures. On his return he 
directed the preparations for the defence of the city, but soon 
decided to become a Muslim. Perhaps the easy terms granted to 
Hawazin had shown him that acceptance of Islam was more profit- 
able than being constantly harried by the former allies whom 
Muhammad had defeated so thoroughly. There was also the 
example of Quraysh to inspire him. Perhaps he wanted to forestall 
B. Malik in gaining Muhammad's favour. Perhaps he hoped to get 
himself recognized by Muhammad as first man in at-Ta'if. Cer- 
tainly he was anxious that the whole city should go over. He re- 
turned from Medina to work for this end, but there was strong 
opposition and he was shot by a fellow citizen and died. The man 
who killed him is sometimes said to have been of B. Malik and 
sometimes of the Ahl^f ; if the latter, the slayer was doubtless of one 
of the other clans of the Ahlaf there is some evidence that each 
of the main groups was divided into smaller groups at variance 
with one another. The latter point is in part confirmed by the flight 
of two prominent kinsmen of 'Urwah's to Medina. 

The situation of at-Ta'if became desperate, however. They were 
isolated, since the allies of both factions had joined Muhammad; 
and the attacks of Hawazin made it difficult for them to leave their 
stronghold. It was an astute move on Muhammad's part to allow 
Hawazin to do his work for him in blockading Thaqif , while at the 
same time the rift between the two tribes became wider and a future 
combination of them against him more unlikely. The initiative was 
taken by two men of the Ahlaf, 'Amr b. Umayyah of the clan of 
'Ilaj and 'Abd Yalil b. 'Amr, who had not been on good terms with 
one another but were brought together by the urgency of the 
position. It was decided to send to Muhammad a deputation consist- 
ing of three men from each of the two main parties ; 'Abd Yalil was 
the leader. In this way it was hoped to avoid the suspicion that any 
individual or family was aiming at the control of the city a sus- 
picion which must, rightly or wrongly, have been attached to 
'Urwah. Muhammad insisted on the destruction without delay of 
the goddess, the Lady of at-Ta'if, al-Lat, but agreed that the task 
might be performed by al-Mughirah and Abu Sufyan. He refused, 
too, to give any dispensation from observing the Worship or ritual 
prayers, and from avoiding usury, wine-drinking, and extra-marital 
relations with women. He seems, however, to have permitted a slight 
relaxation of the hours of the fast of Ramadan. The deputation 


eventually said they would try to induce the rest of Thaqlf to accept 
these terms. This they are said to have done by pretending that 
they had refused the conditions as too onerous; then, when the 
hopelessness of preparing to resist had sunk into the consciousness 
of their fellow citizens, they revealed that they had accepted. 

One curious feature is that Muhammad appointed as leader, 
primarily in the Worship, but perhaps in other respects also, the 
youngest member of the deputation, 'Uthman b. al-'As, of the clan 
of Yasar of B. Malik. He is said to have been the keenest Muslim 
of the party, but perhaps Muhammad also had an eye on the 
relative position of the parties in at-Ta'if ; through al-Mughirah the 
Ahlaf had an influential place in the Islamic state, and it would 
not do to make B. Malik inferior. Another curious point is that 
there is no mention of Thaqlf having to pay the zakdt or legal alms 
among Muhammad's stipulations; and, so far as I have noticed, 
there is no mention of anyone being commissioned to collect any 
contribution or tax from Thaqlf. This might be a reason for the 
disappearance of the text of the treaty with at-Ta'if. Such treat- 
ment of Thaqlf would be in line with Muhammad's generosity to 
Hawazin. While he specially wanted to have Quraysh on his side, 
he also wanted to win over the other Arabs. In cutting short the 
siege of at-Ta'if he was probably trying to avoid bloodshed and 
harsh measures which would exacerbate feelings and make recon- 
ciliation difficult; once it was clear that Thaqlf were not ripe for 
surrender, such a decision was most in keeping with his strategy, 
even if his army had been more suited than it was for siege 

There is no whisper in the sources of any disaffection among 
Thaqlf during the Riddah. Perhaps, however, their attitude was 
not different from that of Hawazin, who had stopped paying their 
legal alms. If Thaqlf had none to pay, then without taking any 
overt action their attitude might well be one of 'wait and see'. In 
view of their great reluctance to accept Muhammad's terms, it is 
plausible to suppose that this was their motive rather than devout 
attachment to the Islamic faith. On the other hand, in the two 
years between the submission of Thaqlf and Muhammad's death 
convinced supporters of the religio-political system of Islam like 
al-Mughirah and 'Uthman b. al- f As may have won a large following 
in the town. Even if this is so, however, the old attachment to the 
rest of Hawazin and consequent opposition to Asad and Ghatafan 


probably played a large part in keeping them from military 
adventures. 1 


A study of the numbers involved in the various expeditions in 
the direction of Syria shows that the road north had a prominent 
place in Muhammad's strategic thinking. 2 As early as August 626 
(iii/5) he led 1,000 men to Dumat al-Jandal, a larger number than 
he had hitherto collected, apart from the visit to Badr in April of 
that year (xi/4) when he had had 1,500. Again, the 3,000 of the 
expedition to Mu'tah in September 629 (v/8) was more than had 
ever gone on any distant expedition and equivalent to the number 
Muhammad commanded at the siege of Medina ; while the 30,000 
of the expedition to'Tabuk (October-December 630 = vii-ix/g), 
even if it is exaggerated, is far more than went on any other expedi- 
tion during Muhammad's lifetime. Unfortunately the sources, 
after giving us these suggestive figures, are tantalizingly reticent 
about details ; and we are left to deduce from our general knowledge 
the reasons for this emphasis in strategy on the road to Syria. 

It would be unrealistic to suppose that Muhammad foresaw the 
later expansion of the Arabs in detail, and indeed no claim of this 
sort is made by the early Muslim sources. As has been hinted in 
previous chapters, however, there were factors within his ken 
which led him in this direction. From at least soon after Uhud he 
seems to have been aspiring to become leader of all the Arabs. 
The Arabs, however, were constantly fighting one another, and 
this fighting helped to keep the population sufficiently small for 
the meagre resources of the desert to support. To keep the Arabs 
under his rule, he must stop inter-tribal fighting; but to do so, it 
was not enough to insist on the acceptance of blood-money instead 
of taking a life for a life ; he must also provide some outlet for the 
warlike energies of the Arabs and for their excess population. This 
outlet he believed was to be found along the route to the north. 
Life in Byzantine Syria must always have seemed infinitely superior 
in material comforts to that of the desert or even of a town like 

1 Akin to Hawazin was Bahilah. It is mentioned as having helped 'Amir b. 
a'sa'ah and Tamim (Caussin de Perceval, ii. 467, 583). Two groups entered 
into agreement with Muhammad after the conquest of Mecca (IS, i/2. 33 ( 61), 
49 .( 93) * c f* Caetani, ii/i. 221-3). The second group was exempted from 
payments. 2 Cf. Excursus B. 


Mecca. Perhaps there was a tradition of preying on the settled 
lands of the empire, though the buffer princedoms had tried to 
stop it. Certainly there would be no difficulty in moving the Arabs 
in this direction provided they were convinced that there was 
a reasonable hope of success. Whether Muhammad was aware of 
the weakness of the Byzantine and Persian empires is a matter of 
conjecture. What he must have realized, however, before his death 
is that the Islamic state was now strong enough to detach the 
border tribes from the Byzantines, and that, once this was effected, 
the settled lands were open to Muslim raiders. He could not have 
been certain that the Arabs would be superior to the Byzantines 
in a pitched battle, though he must have known of the success of 
Arabs against the Persians at Dhu Qar a few years before this. Alto- 
gether he had solid reasons for a policy of expansion northwards. 

In the case of the tribes nearer to Medina, it was necessary to 
pay attention to their relations to one another, but as one went 
north the primary question became that of a tribe's relations to 
the Byzantines. The chief problem came to be how to make them 
leave the Byzantine allegiance. The solution of this problem was 
rendered more difficult by the fact that most of the pro-Byzantine 
tribes were Christian and may have hesitated about accepting the 
religious aspects of Muhammad's movement. In several cases he 
seems to have entered into an agreement with a tribe without 
requiring that the members should become Muslims. 

The northern policy may have affected relations with some of 
the tribes considered in the previous sections. Thus Muzaynah 
seems to have been immediately north of Medina, while part of 
Fazarah was sometimes in Wadi '1-Qura on the usual road to Syria. 
The tribes now to be considered in detail, however, belong pri- 
marily to the northern route. 

Sad Hudhaym and 'Udhrah. According to the genealogists 
'Udhrah was the son of Sa'd Hudhaym, but in Muhamfriad's 
time the names seem to have denoted two distinct tribes. We hear 
of a letter from Muhammad to 'Udhrah being intercepted by a man 
of Sa'd Hudhaym, 1 which suggests that the two groups were not 
on good terms. There are reports of some individuals becoming 
Muslims; the man who intercepted the letter to B. 'Udhrah is said 
to have been killed as a Muslim in an expedition either in 624 (vi/3) 
or 627 (vii/6); another man is mentioned as becoming a Muslim 

1 IS, i/2. 33- 


during the expedition to Tabuk. 1 The so-called 'deputation* to 
Muhammad about the same time probably represented only a small 
section of the tribe. 2 There is also a report of a letter from Muham- 
mad to Sa'd Hudhaym and Judham fixing the sadaqah or 'alms' 
and instructing that this is to be paid to his two commissioners or 
their agents; this presumably belongs to the negotiations with 
Judham late in 627 (vii/6), and will be dealt with presently. All this 
gives the impression that Sa e d Hudhaym was small and weak, and 
closely associated with Judham, or perhaps dependent on it. If we 
may assume that Sa'd Allah is the Islamic form of Sa'd Hudhaym, 
then a section of them received (along with a section of Judham) 
part of the payments made by a Jewish settlement in the north; 3 
this would confirm to some extent the view that they were poor. 

The tribe of 'Udhrah was seemingly more important. In the 
distant past they had helped Quraysh to establish themselves in 
Mecca. One of the early converts from this tribe was a confederate 
of the clan of Zuhrah of Quraysh; 4 and some of Muhammad's 
Medinan followers had mothers from 'Udhrah. 5 Thus the tribe 
was in contact with Mecca and Medina. It appears to have been 
Christian, though the Christianity may have been nominal. 6 The 
account of the 'deputation* to Medina in May /June 630 (ii/9) implies 
that the bulk of the tribe was not Muslim then, and they probably 
did not become Muslim until after Muhammad's death. 7 Some 
individual members of the tribe, however, became attached to 
Muhammad at an early date, and we find him giving them respon- 
sible positions; one commanded the right wing at the battle of 
Mu'tah, and another had oversight of the relations between the 
tribes of Asad and Tayyi'. 8 He also used them as guides. 9 The letter 
already mentioned, that was intercepted by Sa f d Hudhaym, implies 
some understanding with the tribe or a part of it at an early period ; 
and so also does the fact that one of the early converts brought 
sadaqah (Jamrah b. an-Nu'man), and that Muhammad expected 
help from 'Udhrah in an expedition in October 629 (vi/8). 10 

This is the gist of what we can learn about the relations between 
Muhammad and 'Udhrah, except that he made a grant of land to 

1 WW, 401. 2 IS, i/2. 65; Tab. 1722. ii. 

3 WW, 405; cf. Caetani, ii/i. 255. 4 Khalid b. 'Urfutah, IS, iv/2. 74. 

5 Thabit b. Tha'labah, ibid, iii/2. in; 'Amrah bint Sa'd, viii. 271. 

6 Lammens, La Mecque d la Veille de VHtgire, 257/353; cf. 264/360. 

7 IS, i/2. 66 f.; cf. Caetani, ii/i. 229. 8 IH, 793; IS, i/2. 23. 22. 
9 WW, 175, 235- I0 IS, i/2. 74J WW, 315. 


Jamrah and perhaps to another man. 1 The probability is that 
Muhammad had some sort of alliance with the tribe before they 
became Muslims. Even the phrase used about Jamrah is curious : 
'he was chief of f Udhrah, and he was the first of the people of the 
Hijaz who brought to the Prophet the sadaqah of B. 'Udhrah.' 
Could this originally have meant that Jamrah was the first of the 
non-Muslims of the Hijaz to bring a contribution to Muhammad ? 
Or does it distinguish the 'Udhrah of the Hijaz from those to the 
north of it ? If 'Udhrah was receiving letters from Muhammad at 
least by 626, they might have been early in sending sadaqah, even 
though many of them were still not Muslims when the * deputation' 
came in 630. A solution of these problems is conjectural, and is 
bound up with a solution of the general problems involved. 

Judham. The problems concerning Judham are similar to those 
just discussed. Judham, or part of it, was in close relations with the 
Byzantines. The latter used individuals as agents, and employed 
the forces in the defence of the frontier. 2 They are mentioned in the 
sources of Muhammad's life as attached to a Byzantine army; 3 and 
a man of Judham, Farwah b. 'Amr, was commissioner (amil) for 
Caesar in 'Amman and Ma'an, and is said to have become a 
Muslim. 4 Quraysh also seem to have been on good terms with Jud- 
ham, for it was a man of Judham who informed Abu Sufyan 
shortly before the battle of Badr about Muhammad's attempt to 
attack him on the way north. 5 

The first mention of an agreement between Muhammad and 
Judham is about October 627 (vi/6), at or just before the time of 
the expedition of Zayd b. Harithah to Hisma. The story is some- 
thing like this. Dihyah b. Khalifah al-Kalbi, who had gone to Syria 
on an errand for Muhammad, was returning to Medina with gifts, 
when he was robbed by a man of Judham called al-Hunayd. 
Another clan of Judham, however, or some men from another 
tribe, forced al-Hunayd to give the things back. Meanwhile a 
leader of Judham, Rifa'ah b. Zayd, had been in Medina, had 
brought back to the tribe Muhammad's terms for an alliance, and 
the tribe had accepted. Muhammad had not been informed of this 
decision, however, and sent out Zayd b. Harithah to avenge the 
insult to his messenger. There was a skirmish in which the Muslims 

1 IS, i/2. 26. 23, reading 'Udhri for 'Adawi with Usd, s.v. 

2 Lammens, Mecque, 33/129 f.; id., U Arable Occidentals avant VHegire, 
Beirut, 1928, 315 n. 3; cf. IS, vii/2. 148 f.; Usd, iv. 178. 

' IH, 792; WW, 311. * IS, i/2. 18, 31, 83. 5 WW, 40. 


killed al-Hunayd and captured a number of women and animals. 
Rifa'ah was sent to Mecca again to protest, and the dispute was 
settled amicably. So run the sources. 1 The mention of negotiations 
between Muhammad and Judham about April 628 (xii/6), between 
al-Hudaybiyah and Khaybar, 2 must refer to the closing episodes 
of this incident. 

Though the sources say that this agreement involved the accep- 
tance of Islam, it is almost certain that this was not so. The clans 
of Judham who joined together in the transaction included Wa'il, 
and there were also many members of the tribes of Salaman and 
Sa'd Hudhaym. Yet late in 630 on the Tabuk expedition we hear 
of a man of the clan of Wa'il and another of Sa f d Hudhaym being 
converted, and apparently being rewarded for their conversion by 
the gift of some of the revenues from a Jewish settlement. 3 The 
simplest explanation of this is that the rest of Judham had become 
allies of Muhammad without becoming Muslims, and that he was 
now anxious to bring about their conversion. It is difficult to 
suppose that earlier converts from the tribe were less well treated, 
or that gifts to them have been passed over by the sources. 

This hypothesis of alliance without conversion tends to be sup- 
ported by the letter from Muhammad to Rifa'ah which constitutes 
the first part of the account of the 'deputation' of Judham. 4 The 
letter runs: 'This is a letter from Muhammad, the Messenger of 
God, to Rifa'ah b. Zayd and to his tribe and to those who follow with 
them ; he calls them to God ; he who accepts is in the party of God 
(hizb Allah) ; he who refuses has two months' security.' Muhammad 
was surely not in a strong enough position in April 628 to make 
a demand for acceptance of Islam or withdrawal from the sphere 
of Muslim influence. It is not necessary, however, to take the 
words in the way that later practice suggests. The 'party of God* 
might be a 'united front' of Muslims and Cnristians, 5 and the 
letter might be a demand that the members of the group which 
Rifa'ah represented should state definitely to which side they be- 
longed. The incident leading to the expedition of Hisma was 
probably due to some of them joining the group for certain pur- 
poses and not for others, and so trying to have the best of both 
worlds the privileges of an alliance with Muhammad without 
the responsibilities. The story suggests that Judham was at fault, 

1 IH, 975; WW, 235 f. * IS, i/2. 83. i ; cf. Ifdbah, i. 1060. 

3 WW, 405 ; and cf. above. 4 IS, i/2. 82 f. 5 Cf. Q. 5. 56/61 ; 58. 22. 


and Muhammad doubtless wanted to avoid further difficulties by 
having a clear distinction between friends and enemies. 

If the argument so far holds, then the letter to Sa f d Hudhaym 
and Judham about paying sadaqah 1 was sent to this same group. 
The payment may have been in part a punishment for their wrong 
attitude. If not, it must indicate that non-Muslims had to pay for the 
advantages of alliance with Muhammad. The letter seems to refer 
to some specific occasion since the collectors of the sadaqah are 
named ; just before the Tabuk expedition would be a possibility. 

Muhammad's treatment of Judham shows the juxtaposition of 
severity and kindness which characterized much of his activity. 
He could be severe when men were not straight with him and tried 
to 'sit on the fence 1 , but when a man stood out boldly for Islam 
he would be very generous. 

Quda'ah. Strictly speaking, Quda'ah was a large group of tribes 
which included Juhaynah, ' Udhrah, Bali, Bahra', and Kalb. In the 
sources for the life of Muhammad, however, the term appears to 
be used in a more restricted sense, probably in much the same 
way as Sa'd Hudhaym was distinguished from 'Udhrah which was 
formally a part of it. Thus we hear of an expedition against Bali 
and Quda'ah. 2 Shortly before this Quda'ah had been responsible 
for killing Ka'b b. ' Umayr and his party, 3 and there were some men 
of Quda'ah, apparently Christians, among the opponents of the 
Muslims at Mu'tah. 4 

A number of small groups, reckoned by the genealogists as be- 
longing to Quda'ah, may be mentioned here. Jarm, which had been 
involved in the fighting between Bakr and Tamim in pre-Islamic 
times, is reported to have sent a 'deputation* to Muhammad. 5 Al- 
Qayn often in the form Ba'1-Qayn was like many other tribes, 
divided. Some were with the opposing army at Mu'tah, yet Mu- 
hammad expected help from the tribe against Bali and Quda'ah. 6 
In pre-Islamic times a poet of the tribe, Abu 't-Tamahan, had 
been friendly with 'Abdallah b. Jud'an and az-Zubayr b. 'Abd al- 
Muttalib. 7 Salaman sent a 'deputation* to Muhammad in January 
632, and is said to have become Muslim; some had earlier been 
associated with Judham. 8 

1 IS, i/2. 23 f* a IH, 984-6; WW, 315 f- 3 Tab. 1601. 4 WW, 314. 
5 Caussin de Perceval, ii. 582; IS, i/2. 69-71. 6 IH, 792; WW, 315. 

7 Ibn Qutaybah, K. ash-Shi'r wa-sh-Shu'ard\ ed. M. J. De Goeje, Leiden, 
1904, 229 f.; cf. Caussin de Perceval, i. 131; ii. 232. 

8 IS, i/2. 67; WW, 235- 


Bali. Bali, like Judham and other tribes, had members on both 
sides. The commander of the opposing Arab forces at Mu'tah was 
from Bali, 1 and we have just mentioned the expedition shortly after 
Mu'tah against Bali and Quda'ah. On the other hand, there were 
numerous confederates from Bali along with the Medinan Muslims 
at Badr; some clans like Unayf and Marthad had been at Medina 
before the arrival of the Jews and had become confederates of 
these. 2 Moreover, on the expedition against Bali and Quda'ah 
already mentioned, the Muslim leader was 'Amr b. al-'As, whose 
mother was of Bali, and it was hoped that a contingent from Ball 
would join him. When a deputation' came to Medina in June/ July 
630 (iii/9) and professed Islam, they lodged with one of the 
Medinan members of the tribe, already a Muslim. 3 Most interest- 
ing is the letter from Muhammad to the clan of Ju'ayl, in which he 
acknowledges them as part of the clan of c Abd Manaf of Quraysh 
and gives them the 'alms' (sadaqdi) from certain tribes; on these 
terms they profess Islam. 4 This was probably not due to mere re- 
spect for genealogy; the genealogy may rather have been an excuse 
for the generosity. Like similar grants to converts from Sa'd Allah 
(Sa'd Hudhaym) and Judham it suggests that Muhammad realized 
the urgency of establishing his power along the road to the north 
and so securing an outlet for the more turbulent of his followers. 

Bahrc? '. Bahra lived somewhere near Bali, to whom they were 
related. A contingent of them was in the 'Byzantine' army at Mu'tah. 
There is a report of a 'deputation' of thirteen coming to Medina 
and professing Islam, but there is no mention of the rest of the 
tribe doing the same. The story has been preserved in the family 
of al-Miqdad b. 'Amr, who was originally of Bahra' but became 
a confederate and then adopted son of one of the leading men of 
Zuhrah at Mecca. 5 If this is all that can be said in glorification of 
Bahra', one must conclude that very few of them became Muslims 
or even allies of Muhammad. Distance from Medina is perhaps 
the explanation. 

Lakhm. The part of Lakhm with which we are concerned also 
lived about the Syrian border; they were Christians, and co- 
operated with the Byzantines. 6 A letter has been preserved from 

1 IH,79*; WW, 311. 

2 IS, iii/2. 32-37, &c.; as-Samhudi, i. 114 (= Wiistenfeld, 29; see p. 192 
below); cf. also IS, iv/a. 73. 3 IS, i/a. 65 f.; cf. ibid. iv/a. 73, Ruwayfi'. 

4 Ibid. i/a. 24. 5 Ibid. 66; cf. iii/i. 114-16. 

6 IH, 792; WW, 311, 391; Caetani, ii/i. 288 ff. 


Muhammad to the Muslims of Hadas, a section of Lakhm; it 
promises protection to those who are clearly Muslims and warns 
those 'who go back from their religion* that they forfeit this pro- 
tection. 1 This is the attitude of severity which has already been 
remarked on. The case of Tamim ad-Darl and other members of the 
clan of ad-Dar is interesting. Ten of them came to Muhammad 
on his return from the Tabuk expedition with rich gifts and pro- 
fessed Islam. They were not a 'deputation' in the usual sense, how- 
ever, for they did not return to their clan but remained in Medina, 
and were given a yearly allowance of dates from Khaybar. 2 This is 
unusual enough, but there is a curious sequel. Tamim is said to 
have asked Muhammad to give them two villages in Syria, Hibra 
(or Hebron) and (Bayt) 'Aynun, and there is extant the text of a 
letter from Muhammad to Tamim's brother Nu'aym establishing 
their right to these. 3 European scholars have generally agreed that 
the letter cannot be authentic, and a medieval Hanaf I jurisconsult, 
without impugning the authenticity, ventured to hold that even 
Muhammad had no right to give away what was not his to dispose 
of! While we may join in doubting the authenticity of the text, it 
would not be surprising if Muhammad had had some understand- 
ing with them, and indeed had been keeping them for use in the 
penetration of Syria. Another member of Lakhm had proved 
unreliable Hatib b. Abl Balta'ah. He was a confederate of the 
prominent Companion, az-Zubayr, and had fought at Badr and 
carried a letter to the Muqawqis of Egypt for Muhammad; but just 
before the conquest of Mecca he was caught trying to give infor- 
mation to the enemy. 4 Perhaps Muhammad's clemency to him on 
this occasion was with a view to gaining his assistance in dealing 
with Lakhm. 

Ghassdn. The chiefs of the tribe of Ghassan had for long been 
on friendly terms with the Byzantines and in return for a subsidy 
had defended the Byzantine frontier from the nomads. They were 
Christians but supported the monophysites and not the orthodox. 
Relations were disrupted by the Persian invasion of 6 1 3-14, and the 
old arrangements may not have been restored on the Byzantine 
victory in 629.* 

1 IS, i/2. 21. * Ibid, 75; cf. IH, 777; WW, 287; Caetani, I.e. 

3 IS, i/2. 21 ; cf. Caetani, I.e. 
* IS, inli. 80 f., i/2. 16; IH, 809 f.; WW, 325. 

5 Cf. Th. Noldeke, Die Ghassdnischen Filrsten aus dem House Gafna's y Berlin, 
1887, 42 ff. 


There are a number of notices of the passing of messengers 
between Muhammad and various men on the frontier. Dihyah b. 
Khalifah al-Kalbi was sent to the governor ( r aztm) of Bostra with 
a message for Caesar, probably in 627.* Shuja' b. Wahb of Asad b. 
Khuzaymah was sent to (al-Mundhir b.) al-Harith b. Abi Shimr 
of Ghassan, and was badly received; this was perhaps in 628.* A 
letter was sent to Jabalah b. al-Ayham, 'king' of Ghassan, as a result 
of which he is said (wrongly) to have become a Muslim. 3 There is 
a report of ' Ammar b. Yasir being sent to al-Ayham b. an-Nu'man 
of Ghassan, though this may be a variant of the previous one. 4 
Lastly, the expedition of Mu'tah in September 629 (v/8) was to 
punish Shurahbil b. 'Amr of Ghassan for executing al-Harith b. 
'Umayr al-Azdi when carrying a letter from Muhammad to the 
'king' of Bostra. 5 

There is much in these notices that is to be rejected. That the 
letters contained appeals to these men to accept Islam is doubtless 
a later invention, as is the story of Jabalah's conversion. There may 
also be some confusion of names and dates. It seems certain, how- 
ever, that Muhammad was feeling his way by diplomacy towards 
Syria, and that he approached various important persons and used 
a number of messengers. It may well be that at one time, before 
the final victory of Heraclius was certain, Jabalah had a friendly 
understanding with Muhammad. It is only to be expected that in 
these troubled years various leaders would be trying to get the 
better of one another, and that a man's attitude to Muhammad 
might be different at different times. These notices, then, fit in well 
with what we know otherwise of Muhammad's policy of northward 

(It is perhaps worth mentioning some of the dates of the war 
between the Byzantines and the Persians. By 619 the Persians had 
overrun Egypt and all Asia Minor as well as Syria, and were 
encouraging barbarians to ravage the European provinces. From 
622 to 625 Heraclius was campaigning in Asia Minor with some 
success, and in 626 a short siege of Constantinople by the Persians 
and their allies proved a failure. In 627 Heraclius invaded the 
Persian empire, and in December of that year won an important 

1 IS, i/2. 16; cf. IH, 975 f.; WW. 234 f.; Caetani, i. 734. 

2 IS, i/2. 17; cf. Caetani, i. 735. 

3 IS, i/2. 20; cf. Caetani, ii/i. 69 and Noldeke, o.c., 45 f. 

4 .Al-Ya'qubi, Historiae (ed. M. Th. Houtsma), Leiden, 1883, ii. 84; cf. 
Caetani, I.e. 5 WW, 309. 

6783 I 


victory near ancient Nineveh, but had to retreat shortly afterwards. 
In February 628, however, the Persian emperor was assassinated, 
and the son who succeeded him desired peace. By about March 
628 Heraclius could regard himself as victorious, but the negotia- 
tions for the evacuation of the Byzantine empire by the Persians 
were not completed until June 629. In September 629 Heraclius 
entered Constantinople as victor, and in March 630 restored the 
Holy Rood to Jerusalem.) 1 

The story of the deputation' from Ghassan to Muhammad is 
evidence that Ghassan showed no signs of accepting Islam. 2 Even 
if it is true, it amounts to no more than that three unnamed 
members of the tribe came to Muhammad in December 631 (ix/io), 
were convinced of the truth of his claims, but went home and did 
nothing about it; only one lived to make ,1 public profession of 
Islam in 635. Other references show that Ghassan continued to 
oppose the Muslims for some years. 3 Thus Muhammad had no 
success whatsoever in winning Ghassan over to his side. Some 
small settlements may have felt, when they saw the great expedition 
to Tabuk at the end of 630, that Muhammad was a power to be 
reckoned with, but the great majority of the nomads of the Syrian 
border were unconvinced. 

Kalb y &c. The tribe of Kalb had its territory slightly to the 
east of those just considered. This was as much on the route to al- 
c lraq as on that to Syria. 4 Dumat al-Jandal lay within this territory, 
but the settlement itself was in the hands of Ukaydir b. 'Abd al- 
Malik of the clan of as-Sakiin of Kindah. (The main body of 
Kindah lived in the south of Arabia.) An early convert from this 
tribe was Dihyah b. Khalifah, but we are told nothing about 
the circumstances of his conversion or the reason for it, though 
we hear much about his likeness to the angel Gabriel! Muhammad 
used him as an envoy to places on the Byzantine frontier. 5 

Muhammad's expedition to Dumat al-Jandal in August and 
September 626 (iii-iv/5) may have been a punishment for attacks 
on caravans to Medina, 6 or perhaps merely a reconnaissance in 

1 Cf. Ch. Diehl and G. Marcais, Le Monde Oriental de 395 d 1081 (Histoire 
Ge'ne'rale: Histoire du Moyen Age, iii), 2nd ed., Paris, 1944, 144-50. 
* IS, i/2. 71 f.; cf. Caetani, ii/i. 328* 

3 WW, 391 ;cf. Tab. 2081. 

4 For al-'Iraq cf. Tab. 2065; for Syria WW, 175 n. i. 
s IS, iv/i. 184 f., cf. i/2. 16, 28, &c. 

6 Cf. al-Mas'adi, K. at-Tanbih wa'l-hhrdt, Leiden, 1894, 248. 


force; it must have impressed the Arabs, but there is no record of 
any results apart from the capture of some animals. About the end 
of December 627 (viii/6), however, 'Abd ar-Rahman led a force 
to the district, made some sort of agreement with the leader of 
Kalb there, al-Asya* (or al-Asbagh) b. f Amr, and sealed it by marry- 
ing his daughter Tumadir. 1 One account says that al-Asya c became 
a Muslim, but another speaks of 'Abd ar-Rahman collecting jizyah 
or poll-tax, which would imply that he remained a Christian ; the 
latter seems more likely. Finally, about October 630 (vii/9), Khalid 
b. al-Walid was sent from Tabuk to Dumat al-Jandal with 420 
horsemen. By capturing the 'king', Ukaydir b. 'Abd al-Malik al- 
Kindi, he secured the surrender of the stronghold. Apparently an 
immediate payment was to be made of 2,000 camels, 800 slaves, 
400 coats of mail, and 400 lances, while for the future there was 
to be an annual jizyah or poll-tax. 2 

The letters to parts of Kalb and to Ukaydir, and the account of 
the 'deputation* from Kalb, add little to our knowledge. 3 It is almost 
certain that not all the persons mentioned became Muslims, but 
even if they did their number is negligible; on the whole Kalb 
did not become Muslim at this period. All that happened was that 
certain groups settled in or near Dumat al-Jandal were forced to 
make payments to Muhammad. Though these are often called jizyah, 
presumably to show that the persons making them were not 
Muslims, the regulations in the letters suggest that the payments 
were a fixed proportion of the herds and of the produce of the 
palms, and thus were formally similar to the zakdt paid by the 
Muslims. This is doubtless a source of the confusion. 

Various settled communities. During the expedition to Tabuk 
agreements were also made with some settled communities in the 
south of Syria. The most important was probably the Christian 
community of Aylah (the biblical Elath and modern 'Aqabah) at 
the head of the Gulf of f Aqabah.* The 'king' of Aylah, Yuhannah 
b. Rubah, came in person to negotiate with Muhammad. The 
tribute for Aylah was fixed at 300 dinars annually. Near Aylah 
was Maqna, a fishing town inhabited by Jews belonging to B. Jan- 
bah ; they were required to pay annually a quarter of their produce 
of fruit and yarn (and perhaps also of fish). 5 Adhruh and Jarba, 

1 WW, 236 f.; cf. Caetani, i. 700 f. For 'al-Astagh' cf. IS, iii/i. 90. 15, &c. 

2 IH, 903 WW, 403-5. 3 IS, i/a. 34, 36, 68 f.; cf. Excursus G. 

4 IH, 902; WW, 405 ; IS, i/a. 28 f. ( 45), 37 ( 74, 75); cf. Caetani, ii/i. 253 ff- 
* WW, 405; IS, i/a. 28 ( 44), 37 f. ( 75). 


apparently near 'Amman, were also inhabited by Jews. According 
to one version the people of Adhruh were to pay 1,000 dinars; 
according to another the people of the two places were to pay loo. 1 
It is clear, then, that by the late autumn of 630 Muhammad had 
adopted the policy suggested by a verse of the Qur'an (9. 29) : 
'Fight against those who ... do not practise the religion of truth, 
of those who have been given the Book (Jews and Christians), until 
they pay thejizyah. . . .' 

The northern policy. Muhammad laid great emphasis on north- 
ward expansion. The motives were doubtless those already sug- 
gested. The review of the tribes does not support the belief that it 
was the extent of his success in the north that encouraged him to 
devote so much attention to this region. On the whole his successes 
were meagre, and when he died most of the tribes were still Chris- 
tian and friendly to the Byzantines. 

To begin with he seems to have been prepared to form alliances 
with Christian tribes. Perhaps he proposed a united front against 
the Persians, and followed this up with propaganda emphasizing 
what Muslims believed that was also held by Christians. Or perhaps 
he tried to gain the support of the Christian Arabs by siding with 
the monophysites among them against the orthodox. His letter to 
Bishop Dughatir is an example of this type of diplomatic approach. 2 
So long as the Byzantine empire looked like breaking up, he prob- 
ably found some men who were prepared to listen sympathetically 
to his envoys. By 630, however, it must have been clear that this 
policy was failing. Muhammad must have heard of the great occa- 
sion at Jerusalem when the Holy Rood was restored, even if he had 
not heard of Heraclius' triumphant entry into Constantinople. 
Apart from this he would find the tribes becoming less sympathetic 
to himself, until one went so far as to murder his messenger. About 
this time, then, his policy towards the Christian tribes changed, 
in accordance with a Qur'anic revelation. It may have changed by 
the time of the expedition to Mu'tah in September 629, but the 
details are obscure and we cannot be certain. It had certainly 
changed before he set out for Tabuk a year later. 

By this new policy non-Muslim tribes were given a choice 
between accepting Islam and paying annual tribute. In either case 
they became members of the Islamic security system. If they 
refused that, they were killed or enslaved. Muhammad had never 

1 WW. 40; : IS, i/2. 17 f. ( 70. 2 Ibid. 28 (S 4-0: cf. Excursus F. no. n. 


tolerated double-dealing; now he made it more than ever necessary 
for a man to declare the side to which he belonged. Those who 
became subject to Muhammad without conversion to Islam seem 
to have been given a heavy tribute in some cases. On the other 
hand, the clans which accepted Islam met with great generosity. 
It is worth noting the contrast between the people of Adhruh and 
the small groups from Sa'd Allah and Judham. The effect of this 
policy in the first place was to make the Christian tribes more eager 
to support the Byzantines until the tide of war was running very 
much against the latter and in favour of the Muslims. Ukaydir led 
an armed revolt, and the tribes, including Judham, are mentioned 
as sending detachments to the Byzantine armies. 1 In the north 
during Muhammad's lifetime there was no great battle comparable 
to the battle of Hun^yn; nevertheless, the war with Syria may be 
said to have already begun. While Muhammad could not have fore- 
seen the subsequent expansion of the Arab empire in detail, his was 
the far-seeing mind which directed the Arabs' attention to the 
strategic importance of Syria for the new Islamic state. 


The tribes living to the south of Mecca differed in various ways 
from all those hitherto considered, and there was a corresponding 
difference in Muhammad's policy with regard to them. One im- 
portant point is that there were hardly any contacts between them 
and the Muslims until after the conquest of Mecca; and another 
is that at this critical period of their history no strong or statesman- 
like leader appeared, for none of the men of whom we read had any 
inherited influence or innate ability comparable to that of Malik b. 
' Awf of Hawazin or even 'Uyaynah b. Hisn of Ghatafan. This lack 
of leadership may be due to a general decadence of the inhabitants 
of the region, 2 though the evidence is too slight to permit of cer- 
tainty. The south-west corner of the Arabian peninsula was its 
most fertile part, owing to the plentiful rains, and had once had a 
flourishing civilization. The traditional accounts of the bursting of 
the dam of Ma'rib 3 must be based on memories of the breakdown 
of the irrigation system on which this civilization was founded, but 
modern scholarship regards this breakdown as a symptom of the 

1 Cf. Tab. 2065, 2081. 

2 Cf. Caetani, ii/i. 661-9, where the events up to Muhammad's death are 
reviewed. * IH, 8; cf. Q. 34. 16/15. 


decline of the civilization and not as its cause. The latter is perhaps 
to be sought in some change of trade-routes. As the standard of 
living deteriorated some of the Arab tribes returned to nomadism, 
and most left the Yemen and made their way northwards. The 
groups with whom Muhammad had to deal frequently contained 
both nomadic and settled members. The nomads seem to have 
been mostly pagan in religion, and the settlers Christian (or occa- 
sionally Jewish). 

As the indigenous civilization weakened, foreign conquerors 
made their appearance. For some fifty years from about 525 
the region was under the Abyssinians. They were succeeded by 
the Persians, and these retained the nominal sovereignty until the 
various districts were incorporated into the Islamic state ; but owing 
to his poor communications with 'Iraq the Persian governor was 
largely dependent on his own resources, and probably had little 
influence except in the neighbourhood of the seat of government 
at San' a'. Indeed, he and the Abna' (literally 'sons', that is, of Persian 
fathers and Arab mothers) merely constituted one of several groups 
contending for power in the Yemen. The presence of the Abyssinians 
and the Persians, however, probably helped to increase the divisive 
tendencies which were always latent. With the exception of al- 
Aswad (to be described presently) there are no signs of any 
attempt to combine against the Persians. 

In the case of the southern tribes it seems best not to deal with 
them tribe by tribe, but to arrange the source material so as to 
illustrate the main aspects of Muhammad's treatment of them. 
Such information as we have about inter-tribal relations in pre- 
Islamic times does not illuminate the early Islamic period, and our 
knowledge of the latter is often fragmentary. Despite the uncer- 
tainty of many details, however, it is possible to form a coherent 
picture of Muhammad's policy. 

The most important feature of this 'southern policy* is the 
extensive use of diplomatic methods. Before the conquest of Mecca, 
expeditions would of course have been impracticable; but even 
after that event Muhammad made no great show of force in the 
south. The largest expedition in this direction had only 400 men. 
While this may in part be due to the weakness of the southern 
tribes, it also indicates that the road to the Yemen and the road to 
Syria had very different roles in Muhammad's strategic thinking. 
In the south there was none of the sense of urgency which charac- 


terized his 'northern policy* ; he seems to have been ready to allow 
matters to mature in their own time. 

What he did was to support certain local factions when they 
were prepared to accept the minimum conditions which he named. 
Thus in a letter to Qays b. Salamah of al-Ju'fi Muhammad says 
that he has made him agent (istctmaltu-kd) for a group of clans, 
'those who perform the Worship, give the zakat, and contribute 
(saddaqa) of their wealth and purify it' ; this particular plan was 
unsuccessful, for Qays eventually refused to play his part. 1 An 
early instance was that of two men, Artah b. Sharahil and al- 
Juhaysh of the tribe of an-Nakha e , who received Muhammad's 
blessing after explaining that in their tribe there was a group of 
seventy men of a superior class who controlled everything ; Artah 
is said to have led some of the tribe at the conquest of Mecca, and, 
even if these had gone to live in Medina, the leaven of Islam was 
working in the tribe, for the last 'deputation* to visit Muhammad 
was one of 200 men from an-Nakha'. 2 

The tribe of Murad had been defeated by Hamdan in a battle 
about the time of the Hijrah. They had also, perhaps after the 
battle, been in alliance with the 'kings' of Kindah, doubtless as 
inferiors. About 632/10 one of their leaders, Farwah b. Musayk, 
renounced the alliance with Kindah and came to Muhammad. 
After he had been instructed in Islam, Muhammad appointed him 
agent (ista'mala-hu) for the tribes of Murad, Zubayd, and Madhhij, 
and sent him back to his tribe in the company of an early Meccan 
Muslim, Khalid b. Sa'id. 3 This apparently straightforward course 
of events is seen in a new light when we realize that there was 
another faction in Murad whose leader was Qays b. al-Makshuh 
(or, more fully, Qays b. Hubayrah b. e Abd Yaghuth al-Makshuh), 
that Qays was a friend of ' Amr b. Ma'dikarib of Zubayd, and that 
the two took the side of al-Aswad of *Ans and received his support 
against Farwah. 4 Some of the details of these reports may be 
questioned: the alleged conversion of 'Amr b. Ma'dikarib 5 may in 
fact have been no more than a political alliance, though there is 
insufficient justification for regarding the whole incident as an 
invention; and Farwah may have claimed that Muhammad gave 
him a position which he only received later as a reward for his 

1 IS, i/a. 62 ( 106). 2 Ibid. 77 ( 129). 

3 IH, 950 f.; IS, i/a. 63 f. (108). 

4 IH, 951 f.; Tab. 1732-4, I79&, 199^, &c. 

5 IH, ibid.; IS, i/a. 64 ( 109); Tab. 1732-4- 


services against al-Aswad. There is no word of Qays having become 
a Muslim; the appearance of his name in the list of 'provincial 
governors' on the death of Muhammad 1 need only mean that he 
had some share of power in San' a' and that the leading man there, 
Fayruz ad-Daylaml, was on friendly terms with Medina. Whatever 
view we take of the details, it is clear that the progress of Islam 
here was bound up with successful intervention in local quarrels. 

Somewhat similar is the story of al-Ash f ath b. Qays of Kindah 
and Wa'il b. Hujr, a qaylor prince of the Hadramawt. 2 Both claimed 
a certain valley, and Muhammad supported Wa'il; perhaps it was 
to placate al-Ash f ath that he arranged to marry his sister. 3 Not 
surprisingly al-Ash f ath tried to assert his independence of Medina 
on Muhammad's death ; but on the failure of his attempt he became 
a Muslim and was prominent in the conquests. 

In the case of an-Nakha* it was seen that Muhammad apparently 
was ready to support the plebeians against the nobles. It is not 
clear whether there was a noble or patrician class elsewhere. It is 
possible that the * kings' (muliik) of Kindah and Hamdan and the 
'princes' (aqydl, plural of qayl) of Himyar formed a separate class 
or caste, but it is also possible that they were simply the group of 
de facto rulers. They usually had a title beginning with Dhu or 
Lord, such as Dhu '1-Kula', Lord of al-Kula'. Muhammad was 
frequently in contact with these men, and sometimes took the 
initiative. Some came to an agreement with him and were loyal 
during the Riddah. 4 

Sometimes Muhammad encouraged energetic men to use force 
against their neighbours. One was Surad b. 'Abdallah of the tribe 
of Azd Shanu'ah, who came to Muhammad with a dozen or so 
men; Muhammad put him in charge of these men and of any 
others of his tribe whom he could persuade to become Muslims, 
and gave them carte blanche to fight in the name of Islam against 
any non-Muslims in the region. Surad chose to attack a fortified 
place called Jurash; after a month's siege he pretended to retire; 
the besiegers sallied out, hoping to take the withdrawing force at 
a disadvantage, but instead they found Surad prepared for them 

1 Tab. 1983. 

* IS, i/2. 71; cf. Caetani, ii/i. 326 f., with further references. 

3 Cf. p. 397 below, no. 2. 

Cf. IH, 955-7, 963 f.; IS, i/2. 20 ( n, 13), 21 ( 15), 33 ( 5), 64 ( no), 
73 f. ( i24a), 79 f. (133), 84 ( 142) ; Tab. 1989 ; Usd t iv. 147 ; Ifdbah, iii. p. 1004. 
Cf. also IS, iv/2. 115-19, early converts. 


and fled with some loss. Eventually the men of Jurash came to 
make their peace with Muhammad and to accept Islam. 1 

A more important example of such encouragement was Jarir 
b. 'Abdallah of Bajilah. Coming to Muhammad with 150 men he 
accepted Islam. Then, at Muhammad's suggestion, he attacked the 
town of Tabalah and destroyed the idol Dhu '1-Khalasah, which 
was worshipped by Bajilah, the related tribe of Khath'am, and 
others. There was some fierce fighting, and much bloodshed, 
especially among Khath'am, and not long afterwards the heads 
of Khath'am came to offer their submission to Muhammad. 2 Subse- 
quently Muhammad used Jarir as an envoy to two of the Lords, 
and he was with them when Muhammad died. 3 He was prominent 
among the Muslims during the Riddah and afterwards. 

In various ways, then, Muhammad was interfering 'by letter 
and by envoy' in the affairs of the southern tribes. Not merely 
factions within a tribe but whole clans and even tribes were becom- 
ing associated with Medina. This meant that they were being in- 
corporated within what might be called the 'Medinan security 
system' or Pax Islamica. Again and again in the letters we find it 
stated that, if the persons addressed fulfil their obligations, they 
have the covenant or guarantee of security of God and His mes- 
senger (dhimmah, dhimdm)\ this included security for their lives, 
goods, and rights to land. 4 Thus theoretically the whole strength 
of the Islamic state would be exerted against anyone who attacked 
those in alliance with Muhammad. How this was accomplished in 
practice is not clear, since no Medinan troops were stationed in 
the region. We shall not be far wrong, however, in supposing that 
the many envoys whom Muhammad sent to South Arabia, what- 
ever else they may have done, saw to it that the various allies of 
Muhammad helped one another against outsiders and avoided 
quarrels among themselves. By this means Muhammad's guarantee 
of security may be presumed to have been in large measure effective. 

It is worth mentioning here some accounts that have been pre- 
served by at-Tabari. 5 According to these, when Badham, the 
Persian governor, and the people of the Yemen became Muslims, 
Muhammad placed the whole administration of the Yemen under 

1 Ibid. i/2. 71; cf. Caetani, ii/i. 326 f., with further references. 
z IS, i/2. 77 f.; Ibn al-Kalbi. K. al-Asndm, Cairo, 1914/1332, 34-36; IH, 56; 
Tab. 1763. 3 IS, i/2. 20 ( 13); Tab. 1989, letter from Abu Bakr. 

4 Cf. IH, 963 f.; Usd, iv. 147; IS, i/2, passim-, p. 244 below. 

5 1851-3; cf. Caetani, ii/i. 370 f. 


Badham. On the latter's death Muhammad divided the adminis- 
tration among the following persons: 

Shahr b. Badham . . . San'a'. 

'Amir b. Shahr al-Hamdam . Hamdan. 

Abu Musa '1-Ash'ari . . . Ma' rib. 

Khalid b. Sa'id. . . . between Najran, Rima', and 


at-Tahir b. Abi Halah . . 'Akk and Ash'ar. 
Ya'la b. Umayyah . . . al-Janad. 
c Amr b. Hazm .... Najran. 
Ziyad b. Labid . . . Hadramawt. 

'Ukkashah b. Thawr al-Ghawthi as-Sakasik and as-Sakun. 
al-Muhajir b. Abi Umayyah. . Mu'awiyah b. Kindah. 
Mu'adh b. Jabal . . . Teacher of doctrine in the 

Yemen and Hadramawt. 

These reports are not to be accepted without severe criticism, 
but neither are they to be rejected as valueless. The Persian 
governor may not have become a Muslim a point to be discussed 
later and was almost certainly not appointed by Muhammad 
to govern the Yemen in his name. There are no serious objec- 
tions, however, to holding that Muhammad entered into an 
agreement with Badham and recognized him as governor of the 
Yemen; in this way Badham would come within the 'Medinan 
security system*. The reports of what happened on the death of 
Badham appear to be later compilations, but again, though in 
need of criticism, are far from worthless. The men, apart from 
those already in the south, were probably really sent there by 
Muhammad, but on different occasions and with different func- 
tions. Family traditions would preserve the memory of the com- 
missioning, but would be vague about the precise duties to be 
performed. Some later historian would then collect the items of 
information and systematize them according to his understanding 
of the period. Scattered through our sources are many variant and 
supplementary accounts. Thus Khalid b. Sa'id is said to have been 
sent with Farwah b. Musayk of Murad, after his acceptance of 
Islam, in order to supervise the sadaqdt of the tribe. 1 Abu Bakr sent 
al-Muhajir b. Abi Umayyah to the Yemen as a military leader, 2 
and this was possibly his first appearance there; the account of his 

1 IS, i/2. 64. 2 Tab. 1880. 


being sent by Muhammad added that he was ill and could not go 
immediately, and the modern scholar will therefore suspect that 
this is an attempt by his family to glorify themselves by asserting 
that his commission was by the prophet himself and not by his 
successor. Abu Sufyan b. Harb is not mentioned in the lists con- 
sidered, though elsewhere he is said to have been in charge of 
Jurash or Najran. 1 

However much dubiety there may be about some details, there 
is a sufficient body of material to make it reasonably certain that 
Muhammad had a number of agents or residents among the southern 
tribes. In this way, though he could not have had so much influence 
as he had in Medina, he would exercise a measure of control over 
the affairs of the allied tribes and clans, especially their relations 
to one another. In some cases, however, he seems rather to have 
worked through an agreement with the man already in power, such 
as Badham's son Shahr and a chief of Hamdan, 'Amir b. Shahr. 

It also seems clear that these men were agents or residents with 
at most a handful of troops under them. Muhammad seems to 
have calculated that, if any military force was needed, it could be 
supplied by his allies on the spot. Only after the outbreak of dis- 
turbances upon the murder of Shahr b. Badham by al-Aswad 
al- f AnsI was a military commander sent, al-Muhajir, and he did 
not start from Medina with a large army no men at all are 
mentioned but collected his soldiers on the way, first at Mecca 
and at-Ta'if, and then by attaching to himself local leaders like 
Jabir b. 'Abdallah and Farwah b. Musayk. 2 

It may seem contrary to Muhammad's policy of using diplomacy 
in this southern region that he sent out even three expeditions 
against it (in addition to that against Suda' which was prepared but 
not sent owing to the conversion of the tribe in question). 3 These 
expeditions were small, however, and should pernaps be regarded 
as intended to give a slight backing to diplomatic activities. The 
first was of twenty men only against some Khath'am in the neigh- 
bourhood of Tabalah in May/June 630 (ii/9). It can hardly have 
been due to a private grievance, since the leader was Qutbah b. 
'Amir, an early Medinan Muslim and head of the clan of Salimah. 
There is no mention of negotiations, but only of booty captured, 
so it may have been an unsuccessful attempt to destroy the idol 

1 Al-Baladhurl, 59; Usd, v. 216, iii. 12 f.; al-Ya'qubi, ii. 81. 

2 Tab. 1998. 3 IS, i/2. 63 ( 107). 


Dhu '1-Khalasah. 1 The expedition of Khalid b. al-Walid to Najran 
with 400 men in July/August 631 (iii/io) does not seem to have 
involved any fighting, and is said to have resulted in the conversion 
of the tribe. In this expedition, then, the diplomatic activity was 
primary, and the show of force to support that. 2 The third expedi- 
tion was of 300 men in December of the same year, was directed 
against Madh'hij and Zubayd, and was commanded by 'All. After 
a slight skirmish the opponents, who must have been merely a small 
part of Madh'hij, accepted Islam. This is not a clear case of the 
primacy of diplomacy, but that cannot be ruled out, especially 
when it is remembered that in the previous year, according to the 
same sources, Muhammad had put 10,000 men in the field against 
Mecca and taken 30,000 with him to Tabuk. The booty is a diffi- 
culty, but it is unlikely at this period of Muhammad's life that 'All 
would have been interested in raiding for booty. 3 Thus the accounts 
of the expeditions make it clear that Muhammad did not regard 
the south as a suitable sphere for military activity. 

In one or two cases Muhammad may have backed up his diplo- 
matic approaches by economic inducements. Qays b. Malik of the 
clan of Arhab of Hamdan is said to have been made chief and to 
have been given an annual grant of raisins and grain. 4 Again, 
several branches of the tribe of al-Harith b. Ka'b are said to have 
been exempted from the tithe. 5 These facts are reminiscent of 
similar favours given to tribes on the road to the north, apparently 
as a reward for standing out openly as Muslims ; but the informa- 
tion about the southern tribes is so slight that we cannot be sure 
that the preferential treatment of them was for the same reasons 
as that of the northern tribes, especially since there was nothing 
in the south corresponding to the Byzantine empire. Whatever 
Muhammad's reasons in the above instances, his treatment of 
a group from the clan of Ruha' (of the tribe of Madh'hij) is to be 

1 WW, 387. 

* IH, 958-60; WW, 417 n.; IS, i/2. 72 ( 123). The possibility of confusion 
between Khalid b. al-Walid and Khalid b. Sa'fd should not be overlooked. 

3 IH, 967 f., 999; WW, 417-21; contrast Wellhausen, Skizzen und Vorar- 
beiten, Berlin, 1899, vi. 281". and Caetani, ii/i. 323. In IS, i/i. 108. 26 the 
reference is presumably to God. 

4 IS, i/2. 73 ( 1243). In IH, 946 Musaylimah tries to twist complimentary 
remarks by Muhammad into a recognition of his prophethood. 

5 Ibid. 22 ( 22); cf. Caetani, ii/i. 314. Contrast the 'deputation' from Tujib 
(a part of Kindah) which brought fadaqah in A.H. 9 (IS, i/2. 60 f.); Tujib was 
loyal during the Riddah. 


explained differently. This group was granted a hundred loads of 
Khaybar dates annually, but such a generous provision for them 
was doubtless due to their settling in Medina and attaching them- 
selves directly to Muhammad. 1 No names are mentioned in con- 
nexion with this deputation* of the Ruhawlyun, but it is probable 
that their leader was Malik b. Murarah (or Murrah) ar-Ruhawi, 
since he appears in various places as a trusted envoy of Muham- 
mad's to the southern tribes. 2 

Recent European scholars have tended to think that in his 
efforts to gain the support of tribes in the south and elsewhere 
Muhammad was content with purely political alliances and did 
not make any religious demands. Many letters and treaties indeed 
say that the persons concerned are to perform the Worship and 
pay the zakdt\ but these phrases might have been added by later 
editors of the text who, on the basis of their conception of the 
history of Muhammad's lifetime, argued that these conditions must 
have been included. Though there can be no certainty about this 
explaining away of the wording of the texts, it is a ground for not 
basing any argument on these passages. On the other hand, there 
are passages which speak of the destruction of idols. Dhu '1-Kha- 
lasah at Tabalah has already been mentioned. Others who were 
made to destroy their gods on accepting Islam were some men of 
Himyar who worshipped sticks, the tribe of Khawlan, and Dhubab 
of Sa'd al-'Ashirah. 3 In certain cases, then, even if not in all, 
Muhammad seems to have made religious demands. 

An interesting variant is found in the story of Qays b. Salamah 
of the tribe of Ju'fi (a part of Madh'hij), a man whom Muhammad 
had offered to recognize as a chief. 4 As it was the custom of these 
people to avoid eating the heart of animals, Muhammad told them 
that their acceptance of Islam would not be perfect until they had 
eaten some heart. He therefore had a heart brought and roasted 
and despite their fears made them break this pagan taboo. Partly 
because of this demand and partly because the Muslims insisted 
that parents and ancestors who died as pagans were in Hell, Qays 
is said to have broken off relations with Muhammad. 

Some of the passages which have been thought to indicate that 

' Ibid. 76 ( 127). 

2 IH, 956 foot; IS, i/2. 20. 3. He may originally have been sent from the 
south to Muhammad; cf. IH, 955; IS, i/2. 20. 8, 84. 4. 
' Ibid. 32 ( 56), 61 ( 105), 74 ( 4b). 
4 Ibid. 6 1 f. ( 106, first part); cf. p. 119 above. 


Muhammad made agreements without demanding acceptance of 
Islam are inconclusive. In a letter to the clan of Bariq they are to 
give three days' hospitality to needy Muslims passing through 
their lands, and a letter to the family of Dhu Marhab speaks of the 
obligation of all Muslims to help them. The wording has been 
held to contrast the recipients with the Muslims and therefore to 
show that they were not Muslims. 1 The conclusion is not necessary, 
however. Even on the supposition that the recipients are Muslims, 
the letters read naturally; and there is also the possibility that they 
may have been Christians and with Christians, we know, Mu- 
hammad entered into agreements without making religious de- 
mands. These passages are thus no clear evidence of agreements 
with pagans where no religious demands were made. Again, the state- 
ment that at Muhammad's death the two chiefs to whom Jarir b. 
' Abdallah was sent, Dhu 'Amr and Dhu '1-Kula', were not Muslims 
may merely be an inference from the fact that Jarir was still there. 2 

With regard to the southern tribes, then, the case for holding 
that Muhammad was ready to enter into alliances without making 
religious demands is weak. General considerations are also un- 
favourable to such a view. The contacts with the south were mostly 
during the last two years of Muhammad's life. By that time he 
had broken with the Christians in the north, and 'was demanding, 
in return for the advantages of the Pax Islamica, either that they 
became Muslims or that they paid tribute. It is unlikely that he 
was content with something less in the south. Even if th'ere was no 
Byzantine problem there to make him force a decision, there was 
a Persian problem, and there were many Christians. Careful 
scholarship, therefore, cannot sanction abandonment of the view 
of the earliest sources that in general those pagans who entered 
into agreements with Muhammad became Muslims. The standard 
of performance demanded may have been low, but there was at 
least an attempt to root out idol-worship. 

Muhammad's treatment of the Christians of the south seems 
to have resembled what he did in the north; that is to say, they 
were allowed, while remaining Christians, to enter the sphere of 
the Pax Islamica provided they made certain payments, commonly 
referred to as the jizyah or poll-tax. There was probably a great 
difference, however, in the attitude of these southern Christians. 

1 Ibid. 35 ( 70), 81 ( 136), 21 ( 15); Caetani, ii/i. 349, 302. 

* Al-Bukhari, Maghdzi (64), 64; cf. Wellhausen, Skizzen, iv. 106, n. 2* 


In the north Christianity was linked with support of the Byzantine 
empire; but in the south the Christians, after half a century of 
support from the monophysite Christians of Abyssinia (from about 
525 to 575), had fallen into the hands of Persians, and the Persian 
empire was officially Zoroastrian, had political ties with the Jews 
of the Yemen, and tended to support the East Syrian or Nestorian 
form of Christianity against the monophysite. Though the Chris- 
tians of Najran are thought to have been Nestorians by the begin- 
ning of the seventh century, 1 they could not have been so firmly 
attached to Persia as the northern Christians to Byzantium. More- 
over, their negotiations with Muhammad did not begin until 630; 
and in February 628 the Persian emperor had been assassinated 
and the empire, after the exhausting war with the Byzantines, had 
begun to show signs*of collapse. There was thus no strong reason 
why the Christians of the south should not accept any fair offer 
made by Muhammad. 

We have a certain amount of information about the main group 
of Christians, those of the town of Najran (who are themselves also 
sometimes called 'Najran' as if it were a tribal name). They lived 
among the tribe of al-Harith b. Ka'b, most of whom were probably 
pagans, though Muhammad addressed a letter to 'the bishop of 
Banu '1-Harith b. Ka'b and the bishops of Najran'. 2 A 'deputation' 
came to Muhammad from the people of Najran led by the three 
most important men of the community, the 'dqib (or 'lieutenant', 
presumably the civil governor), who was of the tribe of Kindah 
and was called 'Abd al-Masih, the bishop, who was Abu '1-Harith 
b. 'Alqamah of the tribe of Rabi'ah, and a third named as-Sayyid 
b. al-Harith (though as-Sayyid, 'the master', may be a title, and the 
latter part may mean that he belonged to the tribe of al-Harith). 
A treaty of peace was made in which it was agreed that Muhammad 
would not interfere with their ecclesiastical affairs or property, 
that the people of Najran would make an annual payment of 2,000 
garments of stipulated value, and that they would become allies 
of the Muslims and receive protection. 3 In the case of war they 
were to lend the Muslims 30 suits of mail, 30 horses, and 30 camels, 

1 Tor Andrae, Die Ur sprung des Islams und das Christentum, Upsala, 1926, 
ch. i, i (p. 169 f. in Kyrkohist. Arsskrift, 1923). 

2 Cf. Excursus F, no. 16. 

3 JS, i/2, 84 f. ( 143), 21 ( 14), 35 f. ( 72); the two latter are translated in 
Excursus F, nos. 16, 17. Cf. IH, 957. i. 


but there seems to be no question of their taking part themselves 
in fighting. 

From a letter of Muhammad's to some of the local rulers which 
tells them to pay sadaqah andjizyah to his agents, it may be con- 
cluded that there were Christians in the towns who did not become 
Muslims, but we have no means of estimating their numbers. 1 
The instructions which Muhammad is alleged to have given to 
Amr b. Hazm when sending him to Najran state that a Jew or 
Christian who becomes a Muslim is to have the full rights and 
duties of a believer, whereas those who retain their Christianity 
or Judaism are to be left alone but to be liable for zjizyah of a dinar 
per person; for various reasons the chief being the existence of 
a much shorter version it is likely that these instructions have 
been greatly altered and expanded in the course of transmission, 
and now reflect the practice of a later period than Muhammad's 
lifetime. 2 Many certainly remained Christians and were removed 
to 'Iraq in the caliphate of 'Umar b. al-Khattab. Some, however, 
became Muslims; of the members of the ' deputation* from Najran 
the 'dqib and as-Sayyid are said to have returned to Medina shortly 
afterwards and made their profession of faith : and some at least 
of the last deputation' to come to Muhammad, one of 200 men 
from an-Nakha', a branch of Madh'hij, appear to have been 
Christians. 3 

It remains to consider the attitude towards Muhammad of the 
Persian element in the Yemen. This may best be done in connexion 
with the rising of al-Aswad b. Ka'b of 'Ans (a part of the tribe of 
Madh'hij), known as Dhu '1-Khimar, the 'man of the veil'. There 
are discrepancies in the sources, but the following may serve as 
a general account. On the death of the Persian governor in San'a' 
about 631/10 al-Aswad took up arms against his son Shahr, who 
succeeded him, defeated him in battle, and killed him. Shortly 
afterwards he entered San'a'. Though al-Aswad was supreme 
commander, he was in uneasy alliance with various other groups, 
and the leaders of these, notably Qays b. al-Makshuh of Murad and 
Dadhawayh who commanded the men of Persian descent, were 
partly independent and had considerable influence. Within a 
month or two, however, it was clear that al-Aswad was unable to 

1 IS, i/2. 20. 7; cf. IH, I.e. 

2 IH, 961 f.; cf. Abu Yusuf, K. al-Khardj y Bulaq (i88s)/i3O2, 40 f. (tr. E. 
Fagnan, Paris, 1921, 108); and Caetani, ii/i. 317-19. 

IS, i/2. 77 ( 129). 


hold the alliance together. A plot was formed against him, and he 
was assassinated by Qays. These events are usually known as the 
first Riddah in the Yemen, and are distinguished from the second 
Riddah under the leadership of Qays. 1 

It is unlikely that the movement of al-Aswad had as its basis the 
(merely conjectural) local Arab feeling against the Persians, since 
under him, despite the death of Shahr, they were able to maintain 
themselves in San'a' ; only with the rise of Qays do they seem to 
have fled. Indeed the reports say nothing of any ideological foun- 
dation for al-Aswad's movement apart from his employing divina- 
tory and magical practices, and he does not seem to have made 
a serious claim to be a prophet. Thus his activities are best 
interpreted as the attempts of one Arab chief to improve his 
position at the expense of his neighbours. The Persians, who in the 
sources are referred to as 'the sons', al-Abna', had Arab mothers 
and had doubtless become assimilated to the surrounding Arabs, 
so that for most purposes they could be regarded as being on an 
equal footing with the other groups contending for power. What- 
ever support they may have been receiving from Persia before 628, 
after that date there could be no hope of further reinforcements 
or subsidies. The knowledge of the internal weakness of the Persian 
empire may have prompted al-Aswad's attack. 

The sources do not permit us to say whether the Abna' had 
retained something of the official Zoroastrianism, or had adopted 
the Arab paganism of their mothers or even Nestorian Christianity. 
Whatever their religious views, however, they must have been 
perturbed at the downfall of the Persian empire, and ready to 
follow a leader who seemed able to give political and religious 
stability. Thus the Persians, more than anyone else in the Yemen, 
would be open to Muhammad's propaganda. It is therefore not 
improbable that Badham, the Persian governor, and his son entered 
into agreements with Muhammad, though these were doubtless 
of a purely political character. Nor is it improbable that, even 
before the rising of al-Aswad, a prominent member of the Abna', 
Fayruz (b.) ad-Daylami, should have been attracted to Islam. It 
seems to have been immediately obvious that al-Aswad was hostile 
to Islam. At least Muhammad sent an agent or agents to the Abna 1 , 
and these now came over to his side and made a formal profession of 

1 IH, 964; al-Baladhuri, io6f.; Tab. 1745-99, 1853-68; Caetani. ii/i. 672- 
85 ; Wellhausen, Skizzen t vi. 26 ff, 
5783 K 


faith as Muslims. It is difficult, however, to say to what extent 
Muhammad's agents were responsible for the conspiracy which got 
rid of al-Aswad. Despite one or two statements in the sources 
which seem to imply the opposite, Qays b. al-Makshuh probably 
had no understanding with Muhammad, since he is never explicitly 
said to have professed Islam, or even to have visited Medina until 
after his capture by the troops of Abu Bakr. The flight of the 
Abna' from him would thus be due not to anti-Persian feeling on 
his part but to his opposition to Muhammad and Islam. Fayruz fled 
to the tribe of Khawlan, to which he was related, and which 
remained loyal to the Muslim cause throughout the Riddah. 1 The 
statement that al-Muhajir b. Abl Umayyah was sent 'to deal with 
the forces of (al-Aswad) al- f Ansi and to help the Abna' against 
Qays b. al-Makshuh' 2 seems to indicate tht the Abna' had a key 
position in Muslim strategy in the Yemen. 

The death of al-Aswad is said to have taken place a few days 
before that of Muhammad. Qays, with the support of 'Amr b. 
Ma'dikarib of Zubayd, continued the anti-Muslim movement ; but 
the two soon quarrelled with one another and fell into the hands 
of al-Muhajir. This collapse of the second Riddah on the appear- 
ance of a Muslim force that was not large and was perhaps mainly 
from the local tribes, shows that the greater part of the population 
apart from Christians and Jews was now Muslim, while the Chris- 
tians and Jews were in alliance with Muhammad. According to 
modern European ideas the conversions may have been norpinal, 
but the religious element was present ; and, to adapt the words of 
a Muslim tradition, we are unable to split open men's hearts to 
discover how far their belief in Islam was genuine. 


Mahrah. If we now consider the remaining tribes of Arabia 
geographically, starting in the south-east and moving northwards, 
we come first to the tribe of Mahrah. Here two groups seem to have 
become Muslims, 3 but they were probably small, for in the fighting 
of the Riddah we find that there were two factions in the tribe, 
neither of which was Muslim. 4 The persons who became Muslims 

1 Tab. 1991; but cf. al-BaHidhuri, 100 and Caetani, ii/i. 604, expedition 
against Khawlan. 

* Tab. 1880. 3 IS, i/2. 34 ( 67), 83 ( 141). 4 Tab. 1980-2. 


are mentioned in connexion with the fighting. Remoteness from 
Medina doubtless explains the slightness of the contact. 

Azd ' Uman. Remoteness also helps to explain the form of the 
movement towards and away from Islam in 'Uman and al-Bahrayn. 
In 'Uman, where a section of the tribe of Azd was dominant, the 
prince and his brother, Jayfar and 'Abbad (or 'Abd) sons of al- 
Julunda, seem to have approached Muhammad of their own accord. 
He sent f Amr b. al-'As to negotiate with them. Jayfar hesitated 
to relinquish sovereignty to the extent demanded by Muhammad, 
but was eventually persuaded by his brother to accept the terms 
offered. ' Amr thereupon assumed certain judicial functions and 
control of the sadaqah, collecting this last from the rich and giving 
it to the poor. 1 This seems to show that there was social unrest in 
'Uman, though it is difficult to be certain about the precise form 
it took. The growing confusion in Persia may have had something 
to do with it. Even with the measures of c Amr b. al-'As, however, 
and the accession of other groups belonging to Azd 'Uman, 2 the 
Muslim party was weak. On the death of Muhammad, 'Amr 
returned to Medina while Jayfar and 'Abbad took to the moun- 
tains; the anti-Muslim forces perhaps mainly the nomadic ele- 
ment 3 found a leader in Laqlt b. Malik, and were only defeated 
and brought into subjection when a Muslim army from outside 
'Uman was able to join the local Muslims. 4 

'Abd al-Qays. The Muslim party in al-Bahrayn was relatively 
stronger than that in 'Uman, if we may judge from the course of 
the Riddah in these two places. Most of the Arabs of al-Bahrayn 
belonged to the tribe of 'Abd al-Qays, but there were also some 
from Bakr b. Wa'il and Tamim. Moreover the population included 
Persians, Christians, and Jews. A number of separate groups were 
in touch with Medina, and some seem to have taken the lead in 
approaching Muhammad. 5 Unfortunately many of the details that 
have been preserved are obscure. We do not know the relation of 
the settlement of Hajar to that of al-Bahrayn, and thus do not 
understand the relation of the 'master' (sahib) of the one to the 
'master' of the other. The most likely hypothesis is that these two 
and also al-Mundhir b. Sawa, the chief local supporter of Muham- 
mad during his lifetime, were Arab rulers in the Persian interest 

1 IS, i/a. 18 ( 8). * Ibid. 80 ( 134); cf. 23 ( 25), 30 ( 49). 

3 Cf. Wellhausen, Skizzen, vi. 24-26. 4 Tab. 1976-80, &c. 

5 IS, i/a. 19 ( 9), 27 ( 4i, 42), 3* f- ( 57), 54 ( 98); c. Caetani, ii/i. 193- 
206; Wellhausen, Skizzen, vi. 


and with support from Persia. They may have been Christians like 
the king of Hirah who dealt with the Arabs on behalf of the Persian 
emperor; at least al-Jarud b. Mu'alla, who rallied the loyal party 
during the Riddah, had been originally a Christian; 1 and a letter to 
al-Mundhirb. Sawa telling him how to deal with Magiansand Jews 
implies that arrangements had already been made for the Christians, 
and thus suggests that al-Mundhir himself was a Christian. 2 Once 
again it was probably the threatened break-up of the Persian 
empire that caused these men to appeal to Muhammad, even 
though their distance from Medina made it difficult for him to 
give them any military help. 

The 'deputation' from 'Abd al-Qays is said to have come to 
Medina in the year of the conquest of Mecca, 630/8, and is thus 
one of the first 'deputations'. 3 In the two years or so between the 
coming of the 'deputation' and Muhammad's death, things went 
far from smoothly in al-Bahrayn; the pro-Muslim movement had 
its ups and downs, and there even seems to have been some apostasy. 
The absence of military support from Medina must have made 
things difficult for Muhammad's agents in the area, of whom the 
chief was al-'Ala' b. al-Hadrami, a Meccan confederate. The 
difficulties came to a head with the death of al-Mundhir b. Sawa, 
which occurred about the same time as that of Muhammad. 
Muhammad's death probably affected the politics of al-Bahrayn 
only indirectly; with powerful anti-Muslim leaders between them 
and Medina the local anti-Muslim party thought the opportunity 
had come to set up an independent principality. The military leader 
was al-Hutam b. Dubay'ah of the tribe of Bakr b. Wa'il, but the 
plan was to set up a scion of the royal house of al-Hirah as prince 
or king. The Muslim party was rallied by al-Jarud, as already 
mentioned, but it was not strong enough to deal with its opponents 
until al-*Ala' brought an army from outside. 

Hanlfah. More is recorded about the tribe of Hamfah than about 
some other tribes, but it is difficult to wrest a coherent picture 
from the material. The crucial question is the relation to one 
another of the four individuals or groups who were in contact of 
some sort with Muhammad, namely, Hawdhah, Thumamah, the 
members of the 'deputation', and Musaylimah. One of the letters 

1 IS, i/2. 54. 21. * Ibid. 19 (9). 

3 Ibid. 54 ( 98) ; the group who came to Medina for corn in June 625 (i/4) 
were presumably not from al-Baljrayn but from ar-Rawha* near Medina; IH, 
590; WW, 150 f.; cf. ibid. 176 f. 


which Muhammad despatched on his return from al-Hudaybiyah 
was to Hawdhah b. 'All of Hamfah. 1 Hawdhah was possibly the 
strongest man in central Arabia at this time. He was allied to the 
Persians, and was responsible for the safety of their caravans on 
a certain section of the route from the Yemen to Persia. From the 
latter fact it may be inferred that, though most of Hamfah lived by 
agriculture, Hawdhah belonged to the nomadic section of the tribe. 
He was hostile to Tamim, especially to the sub-tribe of Sa'd. In 
religion he was probably a Christian like many members of Ham- 
fah. 2 The sources imply that Muhammad's letter to him was a 
summons to become a Muslim, and he is said to have replied that, 
provided he was given a share in the control of affairs, he would 
become a Muslim. This may or may not have been the real tenor 
of the letters. Hawdkah apparently gave a friendly reception to 
Muhammad's envoy but did not become a Muslim. He is said to 
have died in 630/8. 

Of the second man, Thumamah b. Uthal, stories are told of 
how he was captured by the Muslims in an expedition and won to 
Islam by Muhammad's kind treatment. 3 There seem to be no 
good grounds for denying that he was a Muslim, or at least favour- 
ably inclined towards Islam, by about 631/10. Shortly before his 
death, as portents of the coming storm became visible, Muhammad 
sent out envoys to various friendly leaders ; and one of these envoys 
was to Thumamah. 4 Thumamah thereupon became leader of the 
Muslims among Hamfah and played a useful part in the Riddah. 5 
What is puzzling is that Ibn Hisham speaks of Muhammad writing 
to Thumamah and Hawdhah, 'the two kings of the Yamamah'. 6 If 
they were kings at the same time, but of different sections of the 
tribe, then Thumamah was presumably much inferior in power. 
If, on the other hand, this is taken to imply that Thumamah suc- 
ceeded Hawdhah, then he was far from succeeding to all his 
influence, since most of the tribe followed Musaylimah. It is con- 
ceivable that Thumamah was leader of the nomadic part of the 
tribe and would have inherited Hawdhah's position as Persian 
agent, had Persia not been in disintegration. 

Thirdly, there are the members of the 'deputation' which went 
to Medina and made profession of Islam. 7 The leader was Salma 

1 IS, i/2. 18 ( 7); IH, 971. 2 Caussin de Perceval, ii. 404-8, 575-8. 

3 IH, 996. 4 Tab. 1798. 

5 Ibid. 1910 f., 1916, 1962, 1971. 6 971. 7 IS, i/2. 55 ( 101). 


b. Hanzalah, and one of the members was ar-Rahhal (or Nahar ar- 
Rajjal) b. 'Unfuwah, who was later the most prominent supporter 
of Musaylimah. Musaylimah himself is said to have been one of the 
'deputation', but to have looked after the camels and baggage and 
not to have seen Muhammad; his name was perhaps included in 
the report because an alleged remark about him by Muhammad 
was used by his followers to show that Muhammad regarded him as 
a prophet; 1 it may be assumed, then, that Musaylimah was not 
one of the 'deputation'. We now come to the difficulties. Was the 
deputation sent by Hawdhah or Thumamah ? Or did they come 
of their own accord ? The latter seems most likely, and the visit 
would presumably be after the death of Hawdhah ; in this case 
these were probably the leading men of at least the non-nomadic 
part of the tribe. The situation about 63 1 , -with Hawdhah dead 
and Persia in decline, would incline them to seek support from 
Medina. Everything suggests that at the time of the 'deputation 1 
Musaylimah had not yet set himself up as a prophet, or at least 
had not won any appreciable following. 

In the fourth place comes Musaylimah himself, the so-called 
'false prophet*. About the end of the year 10 (beginning of 632) 
Musaylimah is said to have written to Muhammad as one prophet 
to another, and to have suggested that they divide the land between 
them; Muhammad's alleged reply was a denial of Musaylimah's 
claim to be a prophet. 2 Even if this story has been touched up, 
there may well have been some attempt at negotiation. Certainly 
Musaylimah seems to have come forward before Muhammad's 
death as leader of a political and religious movement. The religious 
aspect appears to have been genuine, and it is natural to suppose 
that Musaylimah had been interested in religious matters for many 
years, and had perhaps been some sort of preacher. It has been 
suggested 3 that Musaylimah was earlier than Muhammad in his 
claim to prophethood. The opponents of Muhammad are said to 
have alleged that he received his revelations from 'a man in the 

1 IS, i/2. 55 ( 101); Tab. 1932, 1941; cf. IH, 946. 

2 IH, 965; IS, i/2. 25 f. (33). 

3 D. G. Margoliouth, 'On the Origin and Import of the Names Muslim and 
IJaniP, Journal of the Royal Asiatic Society, 1903, esp. 485 ff. ; contrast C. J. 
Lyall, 'The Words "Hanif" and "Muslim" ', ibid. 771-84; cf. F. Buhl, art. 
'Musailima* in El (i). There are no grounds for holding that the diminutive 
form is derisive; diminutives were regularly used as ordinary names, e.g. 
Khuwaylid, Khadijah's father. The same holds of Tulayhah. In MfMecca, 29. i, 
the words 'Maslamah or' should be deleted. 


Yamamah called ar-Rahman'; 1 and this presumably refers to 
Musaylimah who, though receiving revelations from ar-Rahman, 
the Merciful (that is, God), was himself called ar-Rahman. 2 
General considerations, however, are against the view that Musay- 
limah was active during Muhammad's Meccan period: had it been 
widely known that he was a prophet, the fact would have been used 
by Muhammad's opponents as an argument against him. It seems 
more likely that the allegation that Muhammad was taught by 'a 
man in the Yamamah called ar-Rahman' and even the attribution of 
the name ar-Rahman to Musaylimah belong to the anti-Muslim 
propaganda of Musaylimah's followers, basing themselves on the 
use of the name in the Qur'an and its special association with the 
Yamamah. 3 Al-Waqidi's story implying that Musaylimah was 
known as ar-Rahman before the Hijrah is doubtful since it contra- 
dicts other reports. 4 It is safest, then, to assume that, whatever 
Musaylimah's past religious practices and experiences may have 
been, he did not attain any wide public notice in his own tribe, still 
less beyond it, until after the death of Hawdhah. 

The Muslim sources, though tending to blacken Musaylimah, 
have preserved some genuine details of his teaching. He employed 
saj\ rhythmic prose with rhyme or assonance, as in the earlier 
passages of the Qur'an. He insisted on uprightness of life, and 
taught the doctrines of resurrection and Divine judgement based 
on what a man has done during his life. Formal prayers three times 
a day and fasting were prescribed. A sanctuary or sacred territory 
was instituted in the Yamamah. 5 Most interesting is a passage in 
which, after an oath which refers to various operations of an agri- 
cultural people such as sowing, reaping, milling, and baking, he 
says, 'You are preferred to the people of the tents (wabar), and 
the people of the villages (madar) are not before you.' 6 The last 
two clauses simply mean 'no one is superior to you', 7 but the oath 
and the following injunction to defend their fields (rtf) seem to 
make it clear that Musaylimah's hearers consisted mainly of 

1 IH, 200. 

2 WW, 58; Tab. 1935. 14; cf. Caetani, ii/i. 641, n. i. 

3 IS, i/i. 1 08. 26, as a name of God presumably. 

4 WW, 58 gives 'Abd ar-RahmSn's original name as 'Abd f Amr, whereas IS, 
ill. 88 gives 'Abd Ka'bah. 

5 Tab. 1916 f., 1930-5; cf. Caetani, ii/i. 636-40. 

6 Tab. 1934. 7 Cf. Lane, s.v. madar (p. 2698). 


From the points just mentioned and some similar ones (like the 
use of the phrase 'kingdom of heaven 1 , mulk a$-samd\ l and from 
the fact that there was much Christianity among Hanifah, it is 
clear that Musaylimah had been largely influenced by Christianity, 
especially by some ascetic trends. If we remember that the rapid 
spread of his movement took place about 630 after the death of 
Hawdhah and when no more help from Persia could be looked for, 
it is further clear that Musaylimah's teaching was attempting to 
provide a religious and intellectual basis for a principality centred 
in the Yamamah and independent of Persia, Byzantium, and 
Medina. A curious regulation to the effect that a man was not to 
have intercourse with any woman so long as he had a son alive was 
perhaps intended to deal with the economic basis of this princi- 
pality; 2 the disappearance of the trade between the Yemen and 
Persia had perhaps affected the Yamamah adversely. 

Musaylimah was thus no mere imitator of Muhammad, since he 
was dealing with a different problem, and in matters of detail was 
possibly more influenced by the local Christianity than by Muham- 
mad. The idea, however, of a state or political system whose head 
was a prophet does seem to have come from Medina; perhaps 
it suggested itself to ar-Rahhal as he meditated on what he had 
seen during his visit to Medina with the 'deputation'. Whether the 
idea first came to ar-Rahhal or to Musaylimah himself, the setting 
on foot of propaganda and the spread of the movement must have 
started soon after the return of the 'deputation*. Some members 
of the 'deputation', though probably not all, abandoned Islani for 
the new movement. We do not know sufficient about it to say 
whether, had it been successful in battle against the Muslims, it 
could have produced an organization of the Arabs comparable to 
that of Islam; there is no mention of the 'holy war' in the teaching 
of Musaylimah, and that was part of the necessary economic basis 
of Arab unity and the ideological ground for Arab expansion. 
Though Musaylimah is said to have been skilful in handling men, 
there is nothing to suggest that he was Muhammad's equal in 
breadth of vision and far-sightedness ; on the other hand, he was 
not unaware of the political realities of the time, and must not be 
regarded as a mere fanatic or visionary. The most serious challenge 
which the nascent caliphate had to face came from Musaylimah's 
movement. So far as concerns the life of Muhammad, however, 

1 Tab. 1917. 2. * Ibid. 4 ff. 


all that needs to be noted is the appearance of this movement and 
its attraction for most of Hanlfah. 

Tamlm. The tribe of Tamlm was scattered over the region 
between the Yamamah and the town of al-Hirah, and some 
branches of it were in close relations with the latter. Many members 
of the tribe were Christians of the East Syrian (or Nestorian) 
church. Though there were settlements in this region, it has been 
suggested that these were peopled by other tribes and that Tamim 
was mainly nomadic. 1 

The first convert from Tamim was al-Aqra c b. Habis, who with 
ten men joined Muhammad on his way to the conquest of Mecca. 
Muhammad treated him with great respect and gave him a hundred 
camels at al-Ji'ranah. Later events give the impression that al-Aqra c 
was not one of the leading men of Tamim; perhaps the respectful 
treatment was intended to make 'Uyaynah b. Hisn of Ghatafan 
jealous or to win over other members of Tamim. The acceptance 
of Islam by al-Aqra' may not have ,been until some time after 
al-Ji'ranah; it is sometimes connected with a curious story about 
a contest of eloquence and poetry to which some men of Tamlm 
challenged Muhammad. 2 

The story of the contest is given separately in Ibn Hisham, but 
al-Waqidi and Ibn Sa'd link it up with an expedition led by 
'Uyaynah b. Hisn. In April 630 (the beginning of the year 9) 
Muhammad sent out men to collect the sadaqat. The envoy to 
B. Ka'b of Khuza'ah was well received by that tribe, but a small 
section of Tamim who lived among them were refractory and re- 
fused to pay, and 'Uyaynah was allowed to go after them and punish 
them; some half-dozen men of Tamim were killed, and about 
fifty men, women, and children taken captive to Medina. A 
'deputation' of important men at Tamim came to ask for their 
release. This incident raises many questions. How did a section of 
Tamim come to be so far from the usual haunts of the tribe ? They 
must have been near Mecca, even if still to the north-east of it. 3 

1 Caetani, ii/i. 218 f., &c.; WW, 386 n.; H. Charles, Le Christianisme des 
arabes nomades sur le Limes et dans le dhert syro-mhopotamien aux alentours de 
VHegire, Paris, 1936, 55, 60 f. 

2 IH, 877-83 (expedition), 933-8 (contest), 985 ; WW, 327, 376, 385 f- J for the 
contest cf. IS, i/a. 40 ( 78) and Usd, i. 119-22 (wrongly so numbered instead 
of 107-10). 

3 Various connexions of Tamlm with Quraysh are recorded: Asma bint 
Mukharribah, the mother of Abu Jahl, 'Ayyash, &c. was of Tamim, and also 
the mother of Firas b. an-Naolr (IS, iv/i. s.v. 'Ayyash, Firas); Abu Jahl had a 


Were they Muslims ? If (as seems likely) they were not, why were 
they liable for the payment of camels to Medina ? Was it because 
they were temporarily reckoned with Khuza'ah, as being under 
their protection? Finally, did the poetical and rhetorical con- 
test take place at the same time as the discussions concerning 
the captives ? 

The answer to the last question is probably in the negative, but 
it is not in itself important. What is important is to know which of 
the leading men of Tamlm became Muslims, if any. Ibn Hisham's 
list of those at the contest is practically identical with the list given 
by al-Waqidi. The most important names are 'Utarid b. Hajib (the 
orator), az-Zibriqan b. Badr (the poet), Qays b. 'Asim, and al-Aqra* 
b. Habis. Of these only the last two appear in Ibn Hisham's list of 
those who came to negotiate about the prisoners ; the others in this 
list are different from those in the previous list, the most influential 
being Sabrah b. 'Amr. Apart from the names it is to be noted that 
while in Ibn Hisham the members of the deputation' at the contest 
become Muslims after it, in al-Waqidi and Ibn Sa'd there is no 
mention of this ; a late writer, Ibn al- Athir, speaks of only al-Aqra' 
making the profession of faith. 1 Were this all, it might be possible 
to harmonize the different accounts as they stand; but another 
piece of evidence forces us to be very sceptical about the conver- 
sions. This is that towards the end of Muhammad's life the persons 
responsible for the sadaqdt of Tamlm were Malik b. Nuwayrah, 
az-Zibriqan b. Badr, Qays b. 'Asim, and perhaps one or two 
others. 2 Now Malik b. Nuwayrah is not mentioned as a member 
of any * deputation' or as having become a Muslim, but it is not 
credible that he should have been omitted had he been present, 
as he seems to have been the leading man of the tribe ; the collectors 
of sadaqdt for Tamlm are indeed no other than the chiefs of the 
various sections. 3 Since Malik thus is almost certainly a non- 
confederate, Yazld b. 'Abdallah, from Tamlm (IH, 509; cf. WW, 82); the father 
of Sa'id b. 'Amr, a confederate of Sahm, was of Tamlm (IS, iv/i, 144). These 
do not explain the presence of some Tamim near Mecca, though they may be the 
outcome of it. 

1 Usd t I.e. 

2 IH, 965; Tab. 1750, 1908 f. 

3 The following are named as leaders in Tab. 1910 f. (Caetani, ii/i. 628 ff.; 
variants in Caussin de Perceval, ii. 461-3): 

Subdivisions of tribe Chief leader Other leaders 

ar-Ribab Pabbah: az-Zibriqan b. Badr 'Abdallah b. afwan 

*Abd Manat: *Imah b. Ubayr 


Muslim, the same must be true of az-Zibriqan and Qays b. 'Asim. 
From this we may conclude that, if they took part in a 'deputation' 
to Medina, they did not become Muslims but only reached some 
understanding with Muhammad, presumably an alliance and an 
agreement that they were to pay sadaqdt. 

What exactly happened during the Riddah among Tamim is 
obscure, partly because one of the chief authorities, Sayf b. ' Umar, 
belonged to Tamim and is thought to have covered up the extent 
of his tribe's apostasy, and partly because the enemies of Khalid b. 
al-Walid have twisted the stories to blacken him. The focus is 
a woman called Sajah who claimed to be a prophetess. Her father 
was of Tamim, but her mother came from the largely Christian tribe 
of Taghlib farther to the north ; Sajah may have lived for some time 
among Taghlib, and in any case was probably a Christian. Her 
claim to receive revelations was probably subsequent to that of 
Musaylimah and Tulayhah, but may have been advanced before 
Muhammad's death. She had military support from followers not 
of Tamim, and at first many of the chiefs of her own tribe, such as 
Malik b. Nuwayrah, were friendly towards her. The main result 
of her appearance, however, was to stir up strife between the 
various subdivisions of Tamim. Some of the parts of which she 
had fallen foul met her private army in battle and defeated it. 
After this her star rapidly waned. She moved towards the Yama- 
mah, perhaps seeking protection rather than a fresh world to con- 
quer, while Tamim soon made its peace with the Muslims. Malik b. 
Nuwayrah was put to death by Khalid b. al-Walid, justly, it would 
seem, for he was the most compromised in the affair of Sajah; the 
other leaders retained their positions. 1 

So far as Muhammad's lifetime is concerned, then, there were 
probably few Muslims from Tamim, and these not the most 

Subdivisions of tribe Chief leader Other leaders 

Sa'd b. 'Awf : az-Zibriqan b. Badr 'Awf b. Bilad 

Zayd Manat (JushamI) 

al-Abna' ,, 

Muqa'is : Qays b. * Aim ( ?)'Amr b. al-Ahtam 

al-Butun: Si'r b. Khufaf 

'Amr b. Bahda Safwn b. Safwan al-Hu?ayn b. Niyar 

Tamim (also over ar-Ribab) 

Khadcjam Sabrah b. 'Amr 

Hanzalah Malik: Waki' b. Malik 


Yarbu': Malik b. Nuwayrah. 

1 Tab. 1908-15, 1925 f., &c.; cf. Caetani, ii/i. 626-35, 651-61. 


important men perhaps al-Aqra e b. Habis, an otherwise unknown 
Sufyan b. al-'Udhayl and his family, and certainly al-Hutat b. 
Yazid, whom Muhammad made a 'brother' of Mu'awiyah b. Abl 
Sufyan. 1 On the other hand, nearly the whole of Tamim seems to 
have entered into alliance with Muhammad and agreed to pay 
sadaqat. This state of affairs is not surprising in view of the swift 
growth of Muhammad's reputation and the weakness of Persia 
and al-Hirah. The fact that the old tribal leaders were responsible 
for collecting the sadaqat confirms the belief that Tamim was not 
Muslim, for this was the usual arrangement with the non-Muslim 
communities in later times ; in this case, however, the sadaqat may 
have been retained for the poor of the tribe. The tribe was not 
necessarily all of one religion. Christianity was probably nearest 
to being the official religion, but acceptance -of it must have been 
largely nominal, and it had done little to modify the outlook and 
ideals of the tribesmen, which were still those of the average 
nomadic Arab . 2 There may have been some Magians (Zoroastrians). 
The presence of a vague monotheism and the absence of idol- 
worship were doubtless felt by Muhammad to justify an alliance 
with this tribe. Their religion, whatever it may have been, did not 
bind them to Persia as the Christianity of the north-western tribes 
made them loyal to the Byzantines. On the contrary they were 
ready to raid the Persian domains, and thus were most suitable 
allies for the Muslims. 

There is no mention of any distinctive teaching by Sajah. Pre- 
sumably the current mainly Christian beliefs of the tribe were 
taken for granted, and guidance given in practical, that is, political 
affairs. It is not impossible that there was an attempt to replace the 
Nestorian doctrines of Tamim by the monophysite doctrines of 
Taghlib ; 3 but we have no information about this. In any case little 
would be said about connexions with other Christians elsewhere, 
since the point of having a prophetess was to be religiously inde- 
pendent. While the changing social situation, with the rise of Medina 
and decline of Persia, favoured the progress of such a movement, 
Sajah does not seem to have attempted to deal with social problems. 
In arguing from the silence of the records we perhaps do in- 
justice to these shadowy figures. Nevertheless the impression we 

' IS,i/2. 4 i(78);IH,933f.,&c. 

2 Cf. fighting within the tribe; also IS, i/2. 56-59 ( 192), esp. 58 foot, 
apparent matriarchy. 3 Cf. Charles, op. cit. 64, 76, &c. 


are given by the sources is that the affair of Sajah was no more than 
a slight variation on the age-old intrigue of the desert. She hoped 
to gain some power for herself; Malik b. Nuwayrah hoped to use 
her to increase his own influence. He is said to have encouraged 
her not to attack the Muslims but to deal with the internal affairs 
of the tribe; he does not seem, however, to have fought for her. 

Bakr b. Wffil and Taghlib. These two related tribes are famous 
in pre-Islamic history for many exploits, and especially for the 
fratricidal war they carried on for many years. By the time of the 
conquest of Mecca they were both largely monophysite Christian, 
at least in name. 1 Bakr had been in alliance with the kings of al- 
Hirah for a time, but they had also been victorious against the 
Persians in the notable battle of Dhu Qar (about 6n) 2 . The forces 
involved in this battlemay not have been large and it may have been 
a skirmish rather than a regular engagement; yet it profoundly 
affected the attitude of the Arab tribes towards the Persians, and 
made them realize that in Persia there was a possible field for raids 
and booty. Some parts of Bakr seem to have lived sufficiently far 
west to be attached to the Byzantines. 3 

Records have been preserved of 'deputations' to Muhammad 
from Bakr and Taghlib, and also from Shayban, an important 
sub-tribe of Bakr, mainly responsible for the victory of Dhu Qar. 4 
Of those mentioned by name in these records, however, none 
appears to have been influential. It is therefore to be concluded 
that no major section of Bakr or Taghlib became Muslim. All the 
more surprising in the light of this conclusion is the appeal of 
al-'Ala' b. al-Hadrami, after the initial defeat of al-Hutam in al- 
Bahrayn, to 'those who remained loyal Muslims' (man aqdma 'aid 
isldmi-hi) of Bakr b. Wa'il to intercept the fugitives. These fugi- 
tives were, of course, of their own tribe ; but only a small fragment 
of Bakr can have been involved with al-Hutam in the Riddah in 
al-Bahrayn. The chiefs named as having been appealed to were 
'Utaybah b. an-Nahhas, 'Amir b. 'Abd al-Aswad, Misma', Khasafah 
at-Taymi, and al-Muthanna b. Harithah ash-Shaybam. 5 These 
were doubtless the chiefs of various sections of Bakr. Probably the 
most important already, and certainly the most important as time 
went on, was al-Muthanna, who played a leading part in the 

1 Charles, op. cit. 3 f.; cf. Wellhausen, Skizzen, iv. 15611.; Caetani, ii/i. 299. 
a Ibid. i. 23? f. 3 WW, 311. 

4 IS, i/2. 31 ( 54), 55 ( 99, 100), 56-59 ( 102); cf. WW, 100, conversion of 
a man from 'Ijl, a branch of Bakr b. Wa'il. * Tab. 1971. 


conquest of 'Iraq. The silence of the earliest sources about conver- 
sion must be taken to imply that these men were not Muslims ; but the 
appeal to them by al-' Ala 1 implies that they were on friendly terms 
with the Muslims. The statement that they 'remained loyal Mus- 
lims' is a misreading of the situation by a later historian; but the 
underlying fact is that they were in alliance with Medina and 
remained loyal to the alliance. 

To go beyond this is to venture into the realms of conjecture, 
with little evidence to guide one. Yet the venture must be made, 
for the question has been raised about the relative positions of the 
Muslims and the eastern tribes in the alliance between them and 
in the advance against Persia. Were the tribes (and al-Muthanna 
as heir of the victors at Dhu Qar) already moving against Persia, 
and did Muhammad (or Khalid after settling the Yamamah) 
humbly ask to be allowed to join them ? Or did the Muslims call 
the attention of al-Muthanna and the others to the possibilities of 
invading Persia? The answer lies somewhere between the two 
extremes. Bakr and Taghlib were in a strong position for bargain- 
ing; they were far from Medina and militarily strong. The initiative 
in forming the alliance must have been taken by the Muslims, 
presumably by Muhammad himself towards the close of his life ; 
and the Muslims seem to have been content with an alliance 
according to which Bakr and Taghlib paid no sadaqah at least 
there is no mention of any payment in the earliest period. It is 
unlikely, however, that Bakr and Taghlib thought of more than 
brief raids on Persia. On the other hand, if what has been said in 
this chapter about Muhammad's northern policy is sound, he had 
been concerned to find an outlet for the energies of the Arabs to 
prevent them rending one another. He had paid most attention to 
the route to Syria, since this was the easiest line of expansion from 
Medina. During the last two years of his life, however, with the 
great increase in the number of tribes in alliance with him or depen- 
dent on him, a second line of expansion towards 'Iraq became 
practicable. Doubtless it was Muhammad who sought alliance with 
Bakr and Taghlib, but it was mainly his strategic conception which 
guided later developments. 


For the historian, contemplating the events described in this 
chapter from his lofty eyrie, it is natural to regard as their most 


prominent feature the large measure of unification of the Arab 
tribes. It does not follow that Muhammad and his advisers saw 
things in this light. The idea that the Arabs constituted a unity 
existed, but only in a rudimentary form. It was through the achieve- 
ments of Muhammad himself that it became more explicitly held. 
The word 'Arabs' is hardly to be found in pre-Islamic poetry, and 
the adjective * Arabic* is said to occur first in the Qur'an. 1 There 
the reference is essentially linguistic: in three passages the clear 
Arabic speech of the Qur'an is contrasted with the indistinct or 
'chewed' speech of the 'barbarian' or foreigner ('ajami); 2 in the re- 
maining passages where 'arabi is used it is an attribute of the 
Qur'an or refers to it. 3 Though there is no word for 'Arabs' in the 
Qur'an and the form al-ardb, which does occur in some later 
passages, means the nomadic tribes as distinct from the Medinans 
and other town-dwellers, the conception of 'Arabs' as a separate 
ethnological or cultural unit is implicit in the use of the word 
'Arabic^TThephrase 'an Arabic Qur r an' indicates that this revela- 
tion is intended for the 'clear-speakers', and the contrast suggested 
to the hearers was doubtless with Abyssinians, Byzantines, Per- 
sians, and perhaps Jews. Thus, at least from about the middle of 
the Meccan period, the religion founded on the Qur'an was re- 
garded as an alternative to any of the religions of these foreigners; 
anti-foreign feeling had much to do with its acceptance in prefer- 
ence to Christianity or Judaism. 

To begin with, Muhammad thought of himself as sent to his own 
tribe (qawm), which presumably means Quraysh; but gradually, 
by steps which are not clearly marked in the Qur'an, he came to 
see his mission as a wider one. Before the Hijrah he had summoned 
some members of nomadic tribes to believe in God, in addition to 
negotiating with the people of Medina.|With the Hijrah the notion 
of an ummah or community with a religious basis became prom- 
inent. The most urgent problem of this community was the estab- 
lishment of peace between the various clans of Medina. This was 
a problem, however, not merely in Medina but throughout Arabia, 
and, as Muhammad showed himself successful in establishing the 
Pax Islamica at Medina and among the surrounding tribes, it was 

1 O. A. Farrukh, Das Bild des Friihislam in der arabischen Dichtung von der 
Higra bis zum Tode des Kalifen 'Umar, Leipzig, 1937, 128. 

2 16. 103/105 EorD; 26. 195, 198 c; 41. 440? 

3 12. 2 c; 20. 113/112 ? D; 39. 28/29E;4i. 32 E + ; 42-7/5 E; 43. 3/2 ? c; 13. 
37 DE (hukm)\ 46. 1 2/1 1 DE (lisdn). 


natural that other tribes would want to take advantage of the new 
system. Muhammad, too, would not be averse to extending his 
security system, since, if details were satisfactorily arranged, exten- 
sion would lead to greater security. Presumably, then, in Muham- 
mad's explicit thought about what he was doing, he conceived 
himself as extending the Islamic community, that is, the body 
of those who professed Islam or who without professing Islam 
believed in God and had placed themselves under His protection 
and that of His messenger. 

The whole of Muhammad's work may be regarded as the build- 
ing on religious foundations of a political, social, and economic 
system; and his tribal policy was merely an aspect of this. The 
Medinan clans which joined with the Emigrants to form the new 
community already had confederates both from among the Jews 
of Medina and from among the surrounding nomads ; and from the 
beginning these confederates shared at least partially in the benefits 
of the new political system and the Pax Islamica. In the early years 
of the Medinan period Muhammad seems to have contracted alli- 
ances with other tribes in the neighbourhood on a purely secular 
basis. Gradually, however, as the sphere of the Pax Islamica became 
wider and Muhammad grew stronger, he began to demand, as 
conditions of alliance, belief in God and recognition of himself as 
prophet. After his disappointment with the nomads who failed to 
join the expedition of al-Hudaybiyah, there was a tightening up, 
and, from this time on, acceptance of Islam presumably afso meant 
acknowledgement of the prophet's right to give orders to all 

It is important to realize that, when Muhammad began to de- 
mand acceptance of Islam from some would-be allies/ he did not 
cease to, make alliances with other groups without any religious 
demand,/' No demand was made of the Meccans when he marched 
into their city in triumph, and many of them took part in the battle 
of Hunayn without being Muslims. The survey of tribes in this 
chapter has shown or suggested that, even up to the time of his 
death and after, there were many alliances with non-Muslims. This 
was normally so in the case of distant and powerful tribes. Though 
such allies were merely secular allies, they belonged in a sense to 
the Pax Islamica in view of current Arab ideas about alliances; 
they shared in its benefits and helped to maintain it. 

As the new social, and political system expanded, Muhammad 


must have given some thought to its economic basis. The problem 
thrust itself upon him with full force during the last two years of 
his life, but he had seen it coming(ln so far as tribes entered into 
the Pax Islamica and stopped raiding one another, the population 
would be larger, since there would be no deaths or other losses in 
raids ; and it would no longer be possible for a tribe, temporarily 
in need, to make good its deficiencies by attacking its neighbour.J 
From the psychological standpoint also, some outlet was required 
for the energies which would otherwise have been spent in the 
razzia. If the Pax Islamica was to be permanent, the standard of 
living must be maintained ; and for that a new source of wealth was 

For a time Muhammad may have looked to increased trade as 
a solution. There was some trade between Medina and Syria, but 
so little is said about it in the sources that it can hardly have been 
important. When Mecca came under his rule, even if its former 
trade had been restored and extended, it would not have been 
sufficient to satisfy the demands of the multitudes who now looked 
to Muhammad as leader. Besides there was the danger that trade 
would foster the false religious attitude that had been the fault of 
the pagan Meccans. So Muhammad felt that trade was not the 

Another possibility was booty from non-Muslims. In the early 
Medinan years this meant a lot to the Muslims, especially the 
Emigrants; just how much it meant to them is difficult to say. It 
was doubtless love of booty that made many men come to Medina 
and attach themselves to Muhammad. In a sense this was the 
solution Muhammad chose, but a further refinement was necessary. 
As the numbers of Muslims grew, and the number of prosperous 
non-Muslims within easy reach decreased, raiding of the traditional 
type became more difficult. If the whole of Arabia were to become 
Muslim, only on the northern frontier would raiding be possible. 
It is one of the great statesmanlike insights of Muhammad that at 
a comparatively early period he conceived of the Pax Islamica as 
embracing all or most of the Arabs, and consequently being forced 
to expand northward. This insight governed his tribal policy. His 
first aim was to see that the members of his community and those 
in alliance with them enjoyed a high degree of security for life and 
property both from enemies without and enemies within. After 
that, however, his chief effort was to increase his influence along 

6788 L 


the road to Syria. In the closing years he also seems to have culti- 
vated the friendship of the tribes in the direction of 'Iraq. In con- 
trast with this he seems to have done little to spread Islam in the 
south and south-east of Arabia. Though he did not refuse any 
prospective Muslims there, he may have looked on them 
more as an embarrassment than an asset more people to pro- 
vide for! 

Such, then, are the general lines of Muhammad's tribal policy. 
It remains to consider the two questions about the extent of 
political relationship to Medina and the relative importance 
of political and religious aspects. Let us start with the second 

If we are to understand the relative importance of religious and 
non-religious motives in the conversion of seventh-century Arabs 
to Islam, we must get rid of the current Western idea that politics 
and religion exist in separate compartments, and we must not 
expect emotional conversions of the type described by William 
James. From the exodus of the Israelites from Egypt, through 
Old Testament times, through New Testament times, through the 
patristic period with its sects, and down to the present day, religion 
and politics in the Middle East have always been closely linked 
with one another. And even if the Muslim had had abnormal 
experiences at his conversion (as some perhaps had), he would have 
lacked words in Arabic to describe his inner states, and would not 
have been sufficiently interested in them to make the effort to 
describe them. 

Tgjajp pmvi'HpH an pmn^fnif^ ^cJalj^nd^^lijdcal systeny the 
Pax Islamica. Of this system religion was an integral part ; it may 
be called the ideological aspect of the system. The peace and 
security given by the system were 'the security of God and of His 
messenger*. We have just seen how, latterly, in appropriate cases 
Muhammad insisted on acceptance of the religious basis. This be- 
came specially necessary in the north-west when the Christian tribes 
there showed clearly that they preferred to remain allies of the 
Byzantines. Now the Islamic system attracted men of the nomadic 
tribes in various ways. It offered an adequate livelihood, mainly 
by booty. It did not involve subjection to a distant potentate; all 
Muslims were in principle equal, and the prophet treated his 
followers with the courtesy and respect shown by a nomadic chief 
to his fellow tribesmen. And when the Byzantine and Persian 


empires showed signs of disintegrating and men needed 'something 
firm to hold on to', the Islamic community promised to have the 
requisite stability. 

The nomads who felt attracted to the new system did not ask 
themselves how far their motives were religious and how far 
secular. For the most part they thought of the system as a whole 
and did not analyse it. They were no doubt ready, however, to 
bargain with Muhammad to try to get him to remit some of the more 
irksome conditions. Contributions to Medina, under whatever 
name, were one disagreeable item. But to some at least it was dis- 
agreeable to acknowledge Muhammad as prophet for the non- 
religious reason that this was tantamount to promising to obey 
him. Thus, apart from details of one's treaty with Muhammad, the 
essential question carjie to be whether to enter the Islamic system 
or to remain outside it. Moreover, after the battle of Hunayn the 
necessity for a decision on this question was thrust upon most of 
the Arabs; and with that came a material reason for accepting Islam 
quickly early application for admission to the system gave some 
advantage over one's rivals. It is not surprising, then, that after the 
conquest of Mecca and victory over Hawazin there should have 
been a * mass-movement' towards Islam. In the religious sphere it 
often happens that seed which has apparently been lying dormant 
in men's hearts suddenly by a change of circumstances finds con- 
ditions suitable for its germination; and when the circumstances 
are common to many men, conversions are widespread. 1 In politics 
there is the familiar phenomenon of 'the rush to get on the band- 

There is thus nothing surprising or impossible about a mass 
movement into the Islamic community in the ninth and tenth years 
of the Hijrah; and consequently there is no justification for reject- 
ing outright the statements in the sources because they tend to 
glorify Muhammad. It may, in European analytical terms, be 
primarily a political movement, but in the integral reality of the 
events the religious and political factors were inseparable. To this 
movement the Riddah was a reaction. It was not the mere revival of 
anything old, whether paganism or pro-Byzantine or pro-Persian 
Christianity. It doubtless had roots in these religious systems, but 

1 Cf. R. Oliver, The Missionary Factor in East Africa, London, 1952, 182-90, 
an instance of how a religious movement with political and other secular conse- 
quences, after a period of slow growth, rapidly expanded in a year or two. 


the reaction of pagan or Christian Arabs to the new circumstances 
created by the growth of the Islamic community produced some- 
thing new. Moreover, as in the movement towards Islam, so in the 
Riddah religious and political factors were inseparably mixed with 
one another. 1 The Muslim historians were therefore right in 
regarding it as a religious movement; it was European scholars 
who erred by taking * religion* in a European and not an Arab 
sense. The Riddah was a movement away from the religious, 
social, economic, and political system of Islam, and so was anti- 

It is worth remarking that, although we speak of the Riddah, 
there were over half a dozen separate movements. One may have 
got some ideas from another, but they were essentially distinct. 
They have their unity from being all parts ofcthe reaction of Arabia 
to Muhammad, but each had its peculiar character. In al-Bahrayn 
and 'Uman there seems to have been little mention of religion; but 
elsewhere the special feature of the Riddah was the appearance of 
'false prophets*, each preaching a new religion with himself as 
centre. Our sources are too meagre for us to be certain about the 
background of these prophets, how far it was pagan and how far 
Christian, how far inspired by Islam and how far a similar but 
independent reaction to similar circumstances. We do not know 
whether their supporters were mainly nomads or mainly agricul- 
turists. If the supporters were settled, then the movements might 
be responses to the challenge from which Islam arose the change 
from a nomadic to a settled economy; if the supporters were 
nomads, the challenge might be the destructive effect of constant 
feuds. The impression given is that only Musaylimah was trying 
to deal with the social and economic problems of his locality ; but 
this may be due to lack of evidence about the others. 

In this diversity the one thing that is clear is that the Riddahs 
were movements of a new type. The appeal was not to something 
old, except at al-Bahrayn, where there was an attempt to restore 
an old dynasty. The inference thus seems to be justified that they 
were reactions to a new situation. Now the new situation might be 
either the rise of a new religious movement in Arabia, or the rise 
of a new political power there at a time when the Byzantine and 

1 The connexion between religion and politics in pie-Islamic Arabia is 
emphasized by J. Ryckmans, U Institution Monarchique en Arabic avant I' Islam, 
Louvain, 1951, 329 ff. . 


Persian empires were in decline. In so far, however, as Islam is a 
religious and political system there is, between these alternatives, 
a distinction but no difference. 

Finally, we have to ask to what extent the tribes were in at least 
political alliance with Muhammad. Those in the neighbourhood of 
Medina and Mecca were all firmly united to him. So also were 
those of the centre and along the route to 'Iraq, but there were 
some exceptions. In the Yemen and the rest of the south-west 
there were numerous groups in alliance, but they may not have 
been more than half the population. In the south-east the propor- 
tion was probably less. Along the route to Syria there had been 
little success in detaching tribes from the Byzantine emperor. 

Thus Muhammad had not altogether succeeded in unifying 
Arabia, but he had clone more than sceptical European scholars 
have allowed. Moreover, his personal influence doubtless gave him 
power and authority beyond that conferred by formal agreements, 
for example, in the affairs of tribes which were in alliance with him 
on an equal footing. There were certainly gaps, but except in the 
north-west they were inconsiderable. The framework of unity had 
been built. A political system with strong foundations had been 
erected, into which the tribes could be brought. Many had come 
in; others could easily be added. The economic basis of the system 
was sound. The quarrels and rivalries of the tribes had not been 
removed, but they had been subdued. Indeed they had been used 
to strengthen the system; the chief motive of tribes like Tayyi 1 
and Hawazin for being loyal during the Riddah may have been 
that their chief rivals were opposed to Medina. 

Religious conversion, as has been seen in our survey, did not 
extend as far as political alliance. It is not easy to find exact details. 
Where a man is said to have made profession of Islam, this was 
probably the case; but, where the sources are silent, it is more 
likely that he did not become a Muslim until later. There must 
always have been a tendency for Christian Arabs in political alli- 
ance with Muhammad or the caliphs to become Muslims. In the 
great upheavals of the age, many men were in need of the support 
given by religion. The 'false prophets' tried to meet this need, but 
had little success. The Christian tribes must have found it difficult 
to 'stand on their own legs' in matters of religion after they had 
been cut off from the Byzantine empire and had seen the Christians 
of 'Iraq suffer from the decline in prestige of the Persian empire. 


For those who had allied themselves politically with Medina, the 
new religion of this rapidly expanding state must have had a great 
fascination and must have seemed the answer to their religious 

JE t' w 4 v-rw /V&C 

needs. -Only a aeeply rooted Christianity could withstand such 



IN Muhammad at Mecca the various clans of Quraysh were named 
and briefly described. Unfortunately, before a similar descrip- 
tion can be given of the social and political groupings at Medina, 
there are various difficulties to be surmounted. 

The first difficulty concerns the reliability of our sources. Those 
for Medina are indeed more ample than those for Mecca; Ibn Sa'd, 
in the second part of his third volume, deals with more than 200 
Medinans who fought at Badr, and in many cases names their 
mothers and wives, while in the eighth volume the women who 
became Muslims to the number of about 400 are similarly 
dealt with. Apart from the children who are also named, we have 
the complete genealogies of perhaps nearly a thousand Medinans 
of Muhammad's time, and know something about their marriages. 
The difficulty is that these genealogies are entirely patrilineal, 
whereas a number of points in the sources make it clear that matri- 
lineal descent counted for something in Medina. Tribes and indi- 
viduals are known as the 'son of such and such a woman'; it is 
noted that a man is the son of the maternal aunt of another man; 
men marry their kinswomen in the female line; and so on. These 
matters will be considered in greater detail subsequently. 1 Our 
information about them is scantier than could be wished, and some 
points are obscure. Sufficient is clear, however, to make us doubt 
whether the neatly arranged patrilineal clans of the later genea- 
logists coincide altogether with the actual social units of Muham- 
mad's time. 

On the other hand, it is also clear that patrilineal descent counted 
for much, and there are no grounds for supposing a general system 
of matrilineal clans. At the time of the Hijrah patrilineal descent 
seems to have been the main principle of organization of the social 
subdivisions of Medina, but within this general framework there 
may have been a number of small groups in which matrilineal 
descent was dominant, although we are seldom able to identify 
these. In practice this means that we accept the patrilineal 

1 Cf. below, p. 378. 


*:.",:':'' I sv a flows 
' modern city 

M Prophets 

B BaqT'aL-Gharqad 

D Zafar.* -.'.;/: 


groupings as generally reliable for Muhammad's time, whatever 
may have been the position earlier, but are on the look-out for indi- 
viduals who tend to side with their mother's kin rather than with 
their father's. As Medina was in a state of transition from nomadic 
practices to those of a settled agricultural community, it is safer to 
assume some lack of homogeneity. 

Even after we have formed a working rule on this point, we have 
still to decide which were the effective groups in the Medinan Arab 
community. There are at least three sets of names to be con- 
sidered. There are the two great tribes, the Aws and the Khazraj ; 
there are the eight clans mentioned in the constitution; and there 
are the thirty-three smaller groups found in Ibn Sa'd's list of the 
Ansar who were at Badr. 

Some doubts may* justifiably be entertained whether the Aws 
and the Khazraj, as important social units, are not the invention 
of the genealogists. For present purposes it is unnecessary to go 
into this question and to try to discover how the genealogists 
manipulated the genuine material which they presumably had. 
It may be that the theory of common descent was invented to 
explain or justify the fact that certain clans usually acted together, 
but it may also have some basis either in blood relationship or in 
religious practice. Yet, even if there really is common descent, this 
was not a strong motive for action in Muhammad's time. There 
was intermarriage between the two alleged tribes, and sometimes 
a clan of one tribe seems to have been more friendly with a clan 
of the other tribe than with the fellow members of its own tribe. 
In the constitution of Medina the groups responsible for blood- 
money are not the two tribes but eight smaller units, and indeed 
the two tribes play no part in that constitution. Moreover, in some 
cases at least the effective units seem to have been smaller than the 
eight clans of the constitution; for example, the constitution speaks 
of an-Nabit, but one of the subdivisions of an-Nabit, B. Harithah, 
was frequently on the opposite side from another, B/Abd al- 
Ash'hal. On the other hand, not all the thirty-three groups of 
Ibn Sa'd were active as independent political units. The accom- 
panying diagrams show the relationships of the clans according 
to Ibn Sa'd, while the map gives their approximate locations. 1 

1 The map is based in part on as-Samhudi's indications and in part on that 
of the Khandaq in Hamidullah, Battlefields of Muhammad, 26 (also the earlier 
version in Bulletin des Etudes Islamiques, 1939, and that in Muhammad al-Qd'id, 
by M. 'Abd al-Fattah Ibrahim). 




V. I 

Qaylah = Harithah 


al-Khazraj (see next 





wf Imru al-Qays Jusham Murrah 

AR . \ KHATMAH 'Amir 

1 1 1 
Silm WAQIF Qays 

Ghanm Zayd 

1 I 
al-#arith ZAFAR 


1 1 1 

. 1 1 

r al-Jaddirah 


I | | AWS (MANAT) 
f Lawdhdn Hubayyib Tha'labah 

Malik Hanash 




, 1 . 
d Mu'dwiyah 


\ayah Umayyah 'Ubayd 







VF Ka'b 


Tha'labah | | 
I al-Khazraj Jusham 

Zayd 'Awf 


Tazid Ghatfb 

Saridah Malik 
Asad 'Abd tfarithah 

'Ali | \ 
Zurayq ftabib 

Sa'd 'Amir 




\ \ 
Ghanm Amr 

\ , \ \ \ 1 J 


| (AL-HUBLA) | 

Ghanm 'Amr 'Amir 



i_ I 

'Adi (B. Mi 

r abdhul), 



(B. IJudaylah) (B. MaghdlaK) 

'AWF clans mentioned in the Constitution. 
SAUMAH clans commonly mentioned as such. 
al-Abjar minor clans or sub-clans. 


(a) The pre-Islamic feuds 

A survey of the pre-Islamic fighting between the clans will give 
some idea of the important units. For many years before the Hijrah 
traditionally for over a hundred, but according to existing 
records for only fifty years or so there had been a series of feuds 
and battles, gradually increasing in numbers involved and in 
ferocity. Originally, like nomadic blood-feuds, they seem to have 
been directed mainly against the persons of the hostile clan and 
possibly against their animals. As time passed, however, the aim 
became more and more the expulsion of the rival group from their 
lands and homesteads, and sometimes even their extermination. 
An economic motive is thus making its appearance, but it is difficult 
to assess its importance accurately. It seems clear that virgin land 
was available even to Umayyad times, 1 but doubtless the labour of 
bringing it under cultivation was considerable and the yield at first 
comparatively poor. Certainly, wherever a clan was strong enough 
to seize old cultivated land it did so, and only those who had been 
expelled from their former lands broke fresh ground. We may 
assume, then, that as the population grew more land was required 
and that the development of new land was a disagreeable task. In 
this sense there was economic pressure. The tendency to self- 
aggrandizement, however, whether in individual or in clan, was 
possibly also not without importance, notably in the career of 
'Amr b. an-Nu'man of Bayadah, about whom more will be said 
presently. The impression one receives is that he was moved, not 
by sheer economic necessity, but by the realization that in extend- 
ing the lands of his clan and of other groups who acknowledged 
him as leader he was increasing his own power. 

The sources separate fighting between an Aws clan and a Khazraj 
clan from fighting between two clans of the same tribe. There is 
little justification, however, for this distinction, and study of the 
events suggests that the conception of two rival tribes was at most 
only being elaborated during this period and had not won general 
acceptance. Fighting was usually between adjacent clans, but the 
fact that neighbours were of the same tribe did not prevent a strong 
clan from attacking them and improving its position at their 
expense. If, as seems to be the case, expropriation of lands from 

1 Cf. J. Wellhausen, 'Medina vor dem Islam' (= Skizzen und Vorarbeiten 
iv/i), 21 n. 


members of the same tribe took place at an earlier date than expro- 
priation from the other tribe, that may be due to the fact that, for 
example, Zurayq (of the Khazraj) when attacked by Bayadah (also 
of the Khazraj) was in a weaker position and less able to gain strong 
allies than 'Amr b. c Awf (of the Aws) when attacked by Bayadah. 1 

The first recorded fighting between the two tribes was in the 
south-west between B. Salim (later identified with Qawaqilah) and 
B. Jahjaba'. Their respective leaders were Malik b. al-'Ajlan (the 
first of the Aws and the Khazraj to assert his independence of the 
Jews) and Uhayhah b. al-Julah, but these two, whose mothers were 
sisters, do not seem personally to have fought one another. In the 
interval between this dispute and the War of Hatib we hear of four 
'wars' : one was between 'Amr b. 'Awf (led by Hudayr b. Simak) 
and al-Harith (led by 'Abdallah b. Ubayy) ; 'in another Mazin put 
to flight Wa'il (led by Abu Qays b. al-Aslat); while in the remain- 
ing two the clans of Zafar and 'Abd al-Ash'hal (the latter under 
Mu'adh b. an-Nu'man) got the better of certain sections of an- 
Najjar. About the same time 'Abd al-Ash'hal was expanding 
northwards and pushing Harithah into fresh lands to the west, and 
Bayadah was expanding at the expense of Zurayq. 

The War of Hatib is the name of a series of incidents which 
culminated in the great battle of Bu'ath shortly before the Hijrah. 

The quarrel began between one Hatib b. Qays (of a branch of 
'Amr b. 'Awf ) and Yazid b. Fus'hum (of al-Harith). Words led to 
blows and then to bloodshed, and others became involved. The 
quarrel may in part be a continuation of the one mentioned in the 
last paragraph, but it is interesting to note that, while 'Amr b. 'Awf 
are still led by Hudayr b. Simak, al-Harith are no longer led by 
'Abdallah b. Ubayy (who belonged not to them but to B. al-Hubla) 
but by 'Amr b. an-Nu'man of Bayadah. It may be conjectured that, 
after the successful expansion of Bayadah against Zurayq, 'Amr 
b. an-Nu'man welcomed an opportunity for further expansionist 
adventures. He was victorious in the first two battles, though in 
the second, in which he had also some of an-Najjar on his side, 
there was great loss of life on both sides. In the next conflict Abu 
Qays b. al-Aslat and 'Abdallah b. Ubayy had joined in on opposite 
sides; Abu Qays seems to have brought with him not merely his 
own clan of Wa'il but the group of clans known as Aws Manat (or 
later Aws Allah); though he was the senior leader on the side of 

1 Contrast ibid. 29 and n. 


'Amr b. 'Awf, he allowed Hudayr b. Simak to command. Hudayr 
was victorious, but, as his side had lost three more men, they 
received three hostages. For reasons not given these hostages 
were killed, and as a result there was some apparently localized, 
but stubborn, fighting between Abu Qays and 'Abdallah b. 

Up to this stage in the war 'Abd al-Ash'hal had apparently not 
been involved, for Hudayr seems to be reckoned to 'Amr b. 'Awf, 
although according to the later patrilineal genealogies he was of 
'Abd al-Ash'hal, and his son Usayd shared the leadership of 'Abd 
al-Ash'hal with Sa'd b. Mu'adh b. an-Nu'man. Now, however, 
the local quarrel of 'Abd al-Ash'hal with their neighbours Salimah 
became linked with the wider quarrel, and they entered on opposite 
sides. In a battle known as the 'day of Mu'abbis and Mudarris' the 
allied clans of the Aws were defeated. 'Amr b. 'Awf and Aws Manat 
made peace, presumably on disadvantageous terms. The curious 
complexity of relationships is shown by the fact that, after Salimah 
had raided the lands of 'Abd al-Ash'hal, the leader of the former, 
'Amr b. al-Jamuh, took under his protection both the person and 
the stronghold of the wounded leader of the latter, Sa'd b. Mu'adh. 
'Abd al-Ash'hal and Zafar, however, refused to submit and left 
Medina (though this possibly means not the whole clan, but only 
the most important men). A deputation went to Mecca, but whether 
with the aim of settling there, or merely in the hope of getting 
military help from Quraysh, is not clear; and in any case they were 
unsuccessful. Eventually they made an alliance with the Jewish 
tribes of Qurayzah and an-Nadir, whose lands some of the best 
in Medina were coveted by 'Amr b. an-Nu'man and Bayadah. 
When 'Amr b. an-Nu'man heard of the intrigues, he demanded 
hostages from Qurayzah and an- Nadir; and when the intrigues 
continued, he had those of the hostages in his immediate power 
put to death. 'Abdallah b. Ubayy, however, disapproved of his 
policy in this matter, and set his hostages free. In the ensuing 
struggle, for which both sides were energetically preparing, he and 
'Amr b. al-Jamuh remained neutral, as did B. Harithah of the Aws. 
At the battle of Bu'ath 'Amr b. an-Nu'man had not merely Baya- 
dah and an-Najjar but also some men from the nomadic tribes of 
Juhaynah and Ashja', while Hudayr b. Simak had a detachment 
from the nomadic Muzaynah as well as the Medinan clans of 'Amr 
b. 'Awf, Aws Manat, 'Abd al-Ash'hal, Zafar, Qurayzah, and 


an-Nadir. The battle was fiercely contested, but at length went 
in favour of Hudayr's side, though both the leaders lost their 

No formal peace was made after Bu'ath, but the combatants 
were too exhausted to continue the struggle actively. For the most 
part the enemy groups avoided one another, but there was a state 
of hostility, and, if a man was careless and gave his opponents an 
opportunity, he was liable to be murdered. This was the un- 
easy position in Medina when negotiations with Muhammad 
commenced. 1 

(b) Description of the individual clans 

Against this background of pre-Islamic history let us try to say 
something about the individual clans as they'were in Muhammad's 
time. The following notes are based mainly on the biographical 
details given by Ibn Sa'd and the geographical information of 
as-Samhudl. 2 

'A bd al-Ash'hal. This clan comes first in the normal order of the 
sources, and it is convenient to describe it first as there is much 
information about it and it is of more than ordinary complexity. 
The leader of the clan in Muhammad's time was Sa'd b. Mu'adh, 
who succeeded to the position of his father, Mu'adh b. an-Nu'man. 
This is one of the very few instances in Medina of an important 
man having a son of comparable importance, and, along with the 
fact that Sa'd and his brother Aws had both married Hind bint 
Simak, suggests that patrilineal descent was more esteemed in this 
clan than in others. With this, however, must be contrasted the 
case of al-Hudayr b. Simak, whose son Usayd appears to have been 
almost as important in the clan about the time of the Hijrah as 
Sa'd b. Mu'adh. It has been seen how al-Hudayr played a leading 
role in Medinan affairs up to his death at Bu'ath, but did so appar- 
ently as the leader of 'Amr b. 'Awf. We know that al-Hudayr's 
sister Hind, who has just been mentioned, had a mother from 'Amr 

1 For the pre-Islamic history of Medina see Ibn al-Athir, al-Kdmil, Cairo 
(i929)/i348, &c., i. 400-20; as-Samhudi, Kitdb WajcC al-Wafd\ Cairo, 
1908-9, esp. i. 152 ff. ; F. Wiistenfeld, Geschichte der Stadt Medina, Gottingen, 
1860 (extracted from Abhandlungen der Koniglichen Gesellschaft der Wissen- 
schaften, ix; it consists of a summary of as-Samhudl) ; Wellhausen, Medina. 

2 IS, iii/a; iv/2. 79-95 (nos. iiSff); viii. 230-337; as-Samhudi, op. cit. i. 
109-16, 134-52; cf. Ibn Durayd, K. al-hhtiqdq ('Genealogisch-etymologisch 
Handbuch'), ed. F. Wustenfeld, Gottingen, 1854, 259 ff. 


b. 'Awf ;' and the same must have been true of al-Hudayr. Presum- 
ably it was the same woman, Umm Jundub bint Rifa'ah b. Zanbar. 
Moreover al-Hudayr seems to have been specially connected with 
those sections of ' Amr b. 'Awf which were settled near Zafar and 
'Abd al-Ash'hal, and latterly at least his utum or stronghold was 
in the territory of 'Abd al-Ash'hal, and he is said to have led 
c Abd al-Ash'hal when they expelled Harithah and occupied their 
lands. As Usayd' smother was also of 'Abd al-Ash'hal, it was natural 
for him to identify himself with that clan. 

The names in the lists of men and women belonging to the clan 
fall into several distinct groups. Apart from what may be regarded 
as the clan proper, there is an important sub-clan which traces its 
descent from a man called Za'ura'. Sometimes the genealogy runs 
Za'ura' b. 'Abd al-Asfhalb. Jusham, sometimes Za'ura' b. Jusham, 
but there is no justification for assuming two distinct persons. 
More primitive appears to be the description of Za'ura' as 'brother 
of 'Abd al-Ash'hal', which is presumably a way of indicating that 
Za'ura' was a group of persons permitted to live alongside and to 
intermarry with the group at that time known as 'Abd al-Ash'hal. 
At one point as-Samhudi says there is some doubt whether Za'ura' 
belonged to the Aws, and at another point he lists a group called 
B. Za'ura' among the Jewish clans, though this may merely mean 
that they were Arabs who had settled in Medina before the Aws 
and the Khazraj and had become subordinated to one of the Jewish 
tribes or clans. 2 The name also occurs in a genealogy of B. 'Adi b. 
an-Najjar of the Khazraj, 3 that of Qays b. as-Sakan, his wife and 
his daughter; and this may indicate another fragment of the 
primitive group. Of those in 'Abd al-Ash'hal one of the chief seems 
to have been Abu '1-Haytham b. at-Tayyihan, 4 but it was his 
mother who was of Za'ura', as according to Ibn Sa'd his father 
was either of Bali or of 'Amr b. Jusham, a group to be mentioned 

Among those mentioned as confederates of 'Abd al-Ash'hal are 
one or two members of the rival clan of Harithah, whose lands 
'Abd al-Ash'hal had seized. We can only guess at the reasons for 
this behaviour, but it is clear that the group were not poor and 

1 IS, viii. 231. 

2 As-Samhudi, i. 136, 115; cf. Wellhausen, Medina, 12; Aghdni, xix. 95; Ibn 
Durayd, op. cit. 263; J. Horovitz, Koranische Untersuchungen, 158. 

3 IS, iii/2. 70. 14-17; viii. 319. 20. 

4 As-Samhudi, i. 136. 


down-trodden, since one of them, Muhammad b. Maslamah, was 
prominent as a Muslim. They all belonged to the sub-clan, B. 
Majda'ah, of Harithah. 

Finally, if we neglect a solitary confederate from the Qawaqilah 
of the Khazraj about whom we have little information, there is 
a curious group known not as 'the sons of so-and-so* but as 'the 
people of Ratij', which is either a stronghold (utum) or a locality. 
As-SamhudI mentions them among the Jewish groups, and Ibn 
Sa'd has some interesting remarks about them. 1 Salamah b. Sala- 
mah of Za'ura' married a woman who was one 'of the Ja'adirah of 
the inhabitants of Ratij of the Aws, confederates of B. Za'ura' b. 
Jusham' ; while of the descendants of f Amr b. Jusham, ostensibly 
a brother of 'Abd al-Ash'hal, it is said that 'these are the people of 
Ratij, except that among the people of Ratij are a group from 
Ghassan from the descendants of 'Ulbah b. Jafnah'. (The Ja'adirah 
are B. Wa'il and some related clans from Aws Manat; B. Jafnah 
appear as a minor group in the Constitution of Medina.) From this 
we see that 'the people of Ratij' were composed of several small 
groups, drawn together for mutual protection, and until shortly 
before the Hijrah probably subordinate to a Jewish clan. 

The genealogies also give us information about the marriages 
of the clan. In roughly half the cases both parties are members of 
the clan, and the different sections enumerated intermarry. About 
half, however, are with other clans, and it is interesting to note 
which these are. The marriages may be divided roughly into earlier 
and later. For the earlier, partners come from al-Abjar, an-Najjar, 
and Sa'idah of the Khazraj and Waqif and 'Amr b. 'Awf of the Aws. 
The later matches are less adventurous, and are confined to their 
allies, Zafar, and to the adjacent Khazraji clans of an-Najjar and 
Salimah. Finally, there is an interesting case ; one woman, Amamah 
bint Bishr, married a Jew of B. Qurayzah, Asad b. 'Ubayd al- 

Altogether this is an illuminating picture of a society in tran- 
sition from a basis of blood to a basis of locality, but it will be best 
to defer remarks of a general nature until the other clans have been 

Harithah. Reference has already been made to the expulsion of 
Harithah from their lands by 'Abd al-Ash'hal. Strictly speaking, 
this was not accomplished by force, but followed on the decision 

1 As-Samhadi, i. 116; IS, iii/2. 16. 6, 21. 10. 2 IS, viii. 236. 


of mediators. The account of as-Samhudi 2 suggests that it was only 
the ufum or stronghold of Musayyir that was forfeited. The clan 
had other strongholds, but Musayyir may have been the main one. 
After a year at Khaybar the main body returned to a site west of 
the later memorial to Hamzah at the battlefield of Uhud. It was 
probably at this time that Muhammad b. Maslamah and others 
became confederates of ' Abd al-Ash'hal, doubtless retaining their 
lands. There is much obscurity, however, about the relations of 
Harithah to e Abd al-Ash'hal. On the one hand, some bitter feeling 
continued; at Bu'ath Harithah refused to fight under al-Hudayr 
b. Simak, who had been chiefly responsible for their expulsion; as 
the Muslims marched to Uhud just before the battle, there was 
nearly a quarrel between Mirba' b. Qayzi of Harithah and Usayd b. 
al-Hudayr, and in theconfusion of the battle it was perhaps not 
altogether an accident that Usayd was wounded by a confederate 
of Harithah. 2 Despite this bitterness, however, the main clan seems 
to have continued to intermarry with the section that had become 
confederates of 'Abd al-Ash'hal and even with some parts of 'Abd 

Apart from this little is to be learnt from the list of genealogies. 
Among the mothers of the men and women who became Muslims 
were women from the clans of an-Najjar, Bayadah, Sa'idah, 
Salimah, Khatmah, and 'Amr b. 'Awf, as well as one from the 
Qawaqilah who were confederates of 'Abd al-Ash'hal. The 
Muslims themselves, however, married within the clan or with 
'Abd al-Ash'hal (including the allied Zafar); only two marriages 
with other clans are recorded Mu'awiyah, a sub-clan of 'Amr 
b. 'Awf, and Mabdhul of an-Najjar. Harithah must have been poor 
after their forced move, and this may have restricted the matches 
open to them. With three clansmen present, they were well 
represented at the convention of 'Aqabah, but only the same 
number at Badr was poor. Among the absentees from Badr was 
al-Bara' b. 'Azib, apparently chief man of the clan after the 

Zafar. The three clans of 'Abd al-Ash'hal, Harithah, and Zafar 
were held to constitute together an-Nabit, but that did not prevent 
the third joining the first in the attack on the second. In general 
Zafar appears to have been dependent on 'Abd al-Ash'hal. Marri- 
ages outside the clan were chiefly with members of an-Najjar, 

1 i. 135-6. 2 WW, 107, 112. 

6783 M 


Salimah and the other two clans of an-Nabit. The leading man 
of the clan under Muhammad was Qatadah b. an-Nu'man. 

'Amr b. 'Azuf. This, like e Abd al-Ash'hal, was a composite group, 
but there is a difference between the two. c Abd al-Ash'hal gives 
the impression of being a young group with centripetal tendencies, 
whereas c Amr b. 'Awf is old and centrifugal. Their lands are 
scattered the sub-clan Jahjaba' is to the west of Quba' in the 
south-west, while Mu'awiyah b. Malik is 'behind' (? east of) Baqi' 
al-Gharqad in the east and consequently they did not have the 
political influence their numbers warranted. 

Of the subdivisions of the clan some appear to have been more 
closely-knit groups than others. Moreover, there are grounds for 
suspecting that in some cases kinship through females (which is 
not fully recorded in our sources) may hav determined the com- 
position of the groups ; e.g. al-Hudayr b. Simak, who led the clan 
although only his mother belonged to it. One such closely-knit 
group is B. Jahjaba'. A generation or two before Muhammad it 
had been strong, for Uhayhah b. al-Julah who belonged to it was 
regarded by some as the leading man among the Arabs of Medina. 1 
A member of this group, however perhaps Uhayhah himself 
was responsible for killing an important member of the clan, Rifa'ah 
b. Zanbar, and B. Jahjaba' are said to have given up at least 
two strongholds as the blood-price, which doubtless weakened 
them considerably. Nevertheless they held together as a group and 
mostly married within the group. Attached to them as confederates 
were some remnants of B. Unayf of the older stratum of Arab in- 
habitants of Medina. 

Another group that appears to have functioned as a group some 
time before the Hijrah was Mu'awiyah, which settled in the east 
beside Baqi' al-Gharqad (later a great Muslim cemetery) and near 
Zafar. There is probably some confusion between this group and 
a similarly named group belonging to an-Najjar. The genealogies 
as given in the sources are quite distinct, but the Western historian 
may wonder whether these are not two fragments of an older group 
which have formed different associations. As-SamhudI mentions 
B. Mu'awiyah among the Arabs in Medina before the coming of 
the Aws and the Khazraj, and also among the Jewish groups. 2 The 
former of these, of course, is quite distinct genealogically from the 

1 Cf. the war of umayr, Ibn al-Athlr, op. cit. 402 f., &c. 


groups among 'Amr b. 'Awf and an-Najjar, but the constructions 
of the later genealogists need not be followed closely. No genealogy 
is given for the * Jewish* group, but they are probably meant to be 
identified with the primitive group. It is stated, however, that they 
lived among B. Umayyah b. Zayd, and these are presumably 
B. Umayyah b. Zayd of 'Amr b. 'Awf, not of Murrah. As Hatib 
b. Qays, from whom the war of Hatib takes its name, is some- 
times attributed to Mu'awiyah and sometimes to Umayyah b. 
Zayd, it seems probable that the 'Jewish' group is to be identified 
with that belonging to 'Amr b. 'Awf. Of the few marriages of 
Mu'awiyah which have been recorded, several are with Harithah. 
To Mu'awiyah also are assigned confederates from Muzaynah and 
Bali. It thus appears to be a mixed group, not unlike B. Za'ura* 
and the 'people of Ratij' in the clan of 'Abd al-Ash'hal. 

Umayyah b. Zayd, which has just been mentioned, is a name 
which also appears in the genealogies as that of a brother of Wa'il 
(and hence a descendant of Murrah b. al-Aws), and one may 
suspect some lost connexion between the two. In 'Amr b. 'Awf 
the group of Umayyah b. Zayd is less closely knit than Jahjaba', 
but still has a definite unity. To it belonged Rifa'ah b. Zanbar, and 
in Muhammad's time his grandson, Abu Lubabah b. 'Abd al- 
Mundhir b. Rifa'ah, was important. If, as is almost certain, the 
mother of al-Hudayr b. Simak was the same as that of his sister 
Hind, then she was a daughter of this Rifa'ah and it was to this 
section of 'Amr b. 'Awf that al-Hudayr was primarily attached. 
To it also belonged Abu 'Amir who went to Mecca rather than 
submit to Muhammad. 

Umayyah b. Zayd is supposed to have had two brothers, 
Dubay'ah and 'Ubayd, and after these also the genealogists name 
groups. Dubay'ah is comparatively well defined. Many of its 
marriages are within the group ; it also married with Waqif and, 
in Islamic times, with the clan of 'Adi of Quraysh, as well as with 
other parts of 'Amr b. 'Awf. Of 'Ubayd little is recorded, though 
Kulthum b. Hidm belonged to it, who was prominent in the early 
days of Islam in Medina. 

There are also one or two less important groups, notably Hanash, 
Hubayyib, Sami'ah (or Lawdhan), and Tha'labah. There is a group 
of confederates, B. al-'Ajlan of Bali, who are attached to Zayd, 
that is, presumably, to Umayyah, Dubay'ah, and 'Ubayd jointly. 
The sub-clan of B. Silm or B. Ghanm b. Silm, to which belonged 


Sa'd b. Khaythamah, though it sprang from a completely different 
branch of the Aws genealogically, had left its kinsmen and joined 
c Amr b. c Awf. 

Thus the clan of 'Amr b. f Awf consists of a number of sections, 
of which some had a distinct existence as groups, while others 
were more nebulous. All the different sections seem to have inter- 
married with one another, to judge from our meagre information. 
They married occasionally with Khatmah and Waqif of Aws 
Manat, with an-Najjar, and with Bayadah and other parts of 
Jusham of the Khazraj. Apart, fyowever, from the marriage at an 
early date of Simak to a woman of 'Amr b. 'Awf, and the marriages 
of B. Mu'awiyah with B. Harithah, we have no record of any 
marriage between 'Amr b. 'Awf and any of the three clans forming 
the Nablt. 1 

Aws Manat Wdqij ', Khatmah , Wffil, fife. The remaining clans 
of the Aws are best considered together. The name Aws Manat 
became in Islamic times Aws Allah, or sometimes, as in the consti- 
tution of Medina, simply al-Aws. There was some dispute about 
the precise application of the name, which is not surprising since 
it is essentially that of the tribe of the Aws. In the constitution, 
however, it appears to mean those clans belonging to the tribe of 
the Aws (according to later genealogists) which we have not yet 
considered. The same group without Waqif and Khatmah was 
apparently called al-Ja'adirah. 

The most important clan, in Islamic times at least, was Khatmah. 
As-Samhudi says that before Islam they were scattered but after- 
wards they gained a centre and multiplied greatly. 2 Ibn Sa'd has 
biographies of fifteen men and women of the clan. Most of the 
marriages recorded are within the clan, but there are also others 
with Waqif, Wa'il, 'Amr b. 'Awf, and Harithah of the Aws, and 
Qawaqilah and al-Harith of the Khazraj. On the whole, however, 
they were not important in the affairs of Medina. 

About the remaining clans we have less information, since with 
one exception Ibn Sa'd has no biographical notes on any members 
of them. It may be that this is because they tended to be opposed 
to Islam, but it is possible that after the battle of Bu'ath they were 
few in numbers. A group of the Ja'adirah whether all or some 
we cannot say had become attached to 'Abd al-Ash'hal. Others 

1 The marriage of ThSbit b. Wadi'ah to a woman of Ratij (presumably after 
the Hijrah) should perhaps be added; IS, iv/2. 86. 2 140. 


may similarly have become more closely connected with the Jewish 
tribes after the battle of Bu'ath, for Aws Manat seem to have been 
interspersed among the Jews. Prior to Bu'ath, Abu Qays b. al- 
Aslat of Wa'il had been one of the leaders of the Aws, senior to 
al-Hudayr b. Simak, but apparently not so influential. When he 
died, less than a year after the Hijrah, he had not become a Muslim, 
though he had been a hanif or monotheist previously and is said 
to have thought of acknowledging Muhammad. Nothing of impor- 
tance is recorded of his son Mihsan. 

It has been suggested that there may be some connexion between 
Umayyah b. Zayd of 'Amr b. 'Awf and Umayyah b. Zayd of Aws 
Manat which the later patrilineal genealogies have obscured. As- 
Samhudl also mentions a small related group called 'Atiyah b. 
Zayd. The general impression given by Aws Manat is that it is 
a heterogeneous collection of old groups whose strength was de- 
clining. They lacked both genealogical and geographical unity, 
though they were all towards the[south of the Medinan oasis. What- 
ever may be the reason, they carried little weight in the Medina 
that Muhammad found. 

An-Najjdr. The most numerous clan or clan-group among the 
Khazraj, and indeed among the Ansar as a whole, was an-Najjar. 
This was an amorphous body, somewhat like 'Amr b. 'Awf, but 
not so scattered. The genealogists arrange the subdivisions under 
the four sons of an-Najjar, who is also known as Taym Allat, later 
Taym Allah. 1 Of these B. Mazin b. an-Najjar seems to have been 
a distinct entity. There is a record of a blood-feud between Mazin 
and Wa'il, in which Abu Qays b. al-Aslat of Wa'il was put to 
flight; 2 and the relative adjective Mazim is frequently used, where- 
as members of the other sections are usually called just Najjari. 
Most of the marriages recorded are within B. Mazin; there are also 
several with other sections of an-Najjar, and only one or two with 
other clans. 

The largest part of an-Najjar was B. Malik b. an-Najjar, but 
within this are several curious groups, B. Hudaylah (or Mu'awiyah), 
B. Maghalah, and B. Mabdhul. Hudaylah and Maghalah were 
women. Few members of these groups are given biographical notices 
by Ibn Sa'd. Of Hudaylah the only man of note is Ubayy b. Ka'b, 
one of Muhammad's secretaries. To Maghalah belonged Hassan 
b. Thabit the poet. Many of the marriages noted are within B. 

1 Wellhausen, Medina, 6. 2 Ibn al-Athir, i. 407 f. 


Malik b. an-Najjar, but very few are restricted to the smaller group. 
There is no obvious peculiarity about the groups named after 
women, and they appear to be based on patrilineal descent at the 
time of the Hijrah. It is curious, however, that there is said to have 
been a dispute whether 'Abdallah b. Ubayy belonged to Ba'l- 
Hubla (as is commonly said) or to B. Maghalah; 1 the only con- 
nexion seems to have been that he married a woman of Maghalah, 
and it therefore looks as if he lived with his wife's group for a time. 

B/ Adi b. an-Najjar were also numerous, but had little distinctive 
character; indeed through intermarriage they were much mixed 
with Malik b. an-Najjar. They included some persons who had 
the name Za'ura' in their genealogy, which suggests some con- 
nexion with the B. Za'ura' of 'Abd al-Ash'hal, though, apart from 
the fact that the unusual name as-Sakan occurs in both groups, 
there is no confirmatory evidence. B. Dinar b. an-Najjar is smaller 
but more compact; that is to say, there is more intermarriage 
within the group ; some of this intermarriage appears to be between 
those who are related on the female side. 

Thus an-Najjar is a clan or clan-group, into which several 
smaller groups are in process of being absorbed so that they cease 
to exist as distinct entities. The large group has as its basis patri- 
lineal kinship, but in some of the smaller groups matrilineal kinship 
may have played a part, though our evidence is insufficient to show 
whether it ever was the main basis of these groups. More and more 
the smaller groups seem to have been intermarrying, and there 
was a slight amount of intermarriage with most of the other clans 
of Medina except those of Aws Manat. The lands of an-Najjar 
became the site of the Muslim city, doubtless owing to the presence 
of the house of Muhammad, which became the central mosque of 
Medina; but prior to the Hijrah the point of greatest density of 
population was probably farther south. Whatever the reason may 
be, an-Najjar, despite its numbers, came behind some other clans 
in political importance at least it did not produce a leader of the 
first rank. As'ad b. Zurarah came nearest to this description, but 
unfortunately died a few months after the Hijrah. After him the 
most important were Mu'adh b. al-Harith and his brothers 
Mu'awwidh and 'Awf, commonly called the sons of 'Afra' 
after their mother. Those just named were all of B. Malik b. 

1 As-SamhudI, 142. 


Al-Harith. The descendants of al-Harith, often known by a 
contracted form Ba'1-Harith, were not a vigorous clan. We hear 
of them ceding a stronghold to a clan of the Aws, probably Waqif, 
and of sections of them going off to Syria. Apart from this they 
present some of the features of an-Najjar, but on a smaller scale. 
One of the small groups which is being absorbed by the clan is 
said to be descended from twin sons of al-Harith, Jusham and 
Zayd. This is doubtless a device to effect the integration of two 
families, or else to explain something of this sort. As the group 
is also known as 'the people of as-Sunh' (a place), its unity is 
possibly based on the fact of common habitation and not on blood. 
Other two small groups are B. al-Jidarah and B. al-Abjar. The 
latter are said to be 'brothers' of the former, and are also known 
as B. Khudrah frorrv the mother of al-Abjar (with the relative 
adjective Khudri). B. al-Abjar, to judge from our scanty informa- 
tion, was a completely exogamous group apparently the only 
example in Medina. Apart from Abu Sa'id al-Khudri, from whom 
many traditions are narrated, no man of note came from B. al- 
Abjar, but the mothers of Sa'd b. Mu'adh and As'ad b. Zurarah 
were sisters belonging to it. Two important men from what may be 
called the main section of the clan were Sa'd b. ar-Rabl* and 
'Abdallah b. Rawahah. 

'Awf: Bctl-Hubld and Qawaqilah. This is one of the points at 
which the fictitious character of the earlier parts of the genealogies 
becomes obvious. Two or three generations before the Hijrah it was 
usual to speak of a clan of Salim, and to this belonged the leader 
who made the Arabs independent of the Jews, Malik b. al-'Ajlan. 
The probability is that there was only a single clan of this name, 
but it seems to have split up into three, and the genealogists have 
conveniently produced three men of the name of Salim. From 
these are descended three clans commonly known as Ba'1-Hubla, 
the Qawaqilah, and Waqif. The two former are most closely con- 
nected, since according to the genealogists both are descended 
from 'Awf b. al-Khazraj ; moreover, in both, Salim is connected 
with a man called Ghanm, but there is doubt about the precise 
relationship. Waqif is one of the clans of Aws Manat, and therefore 
far removed genealogically from B. f Awf of the Khazraj ; but he 
is also known as Salim, and he had a brother Silm (or Salm) who 
had a son Ghanm. Waqif intermarried with both the others; so 
far as our records go, it was the only section of Aws Manat to do so. 


These facts together tend to show that there was a close relation- 
ship between these three clans which the patrilineal genealogies do 
not reveal. 

At the time of the Hijrah Ba'1-Hubla and al-Qawaqilah were 
not greatly different in character from several other small clans. 
The majority of marriages were within the clan or with confeder- 
ates or members of the other clans known as Salim. In addition 
there were a few marriages between Ba'1-Hubla and B. Maghalah 
of an-Najjar, Ba'1-Harith, Sa'idah, Zurayq, and 'Amr b. 'Awf. 
Al-Qawaqilah intermarried with other parts of an-Najjar and with 
Ba'1-Harith, Bayadah, 'Amr b. 'Awf, 'Abd al-Ash'hal, and Zafar. 
These facts link up with others. It has already been noticed that 
a section of al-Qawaqilah how large we cannot tell had become 
confederates of 'Abd al-Ash'hal (with whom ^Zafar were in alliance). 
Again, in pre-Islamic times there was great rivalry between 'Abdal- 
lah b. Ubayy of Ba'1-Hubla and the chief of Bayadah; and it is 
therefore interesting to note that, while al-Qawaqilah intermarries 
with Bayadah, Ba'1-Hubla intermarries with its enemy, Zurayq. 
Ba'1-Hubla was presumably the stronger in view of the influential 
position of 'Abdallah b. Ubayy in Medina, though the latter may 
have been largely due to his personal qualities. The leading man of 
al-Qawaqilah, 'Ubadah b. as-Samit, was not without importance. 
(It may further be noted that there is much disagreement about the 
meaning and origin of the name al-Qawaqilah ; Wellhausen suggests 
that it comes from the name of a place, but the sources look for 
men called Qawqal.) 1 

Sd'idah. The clan of Sa'idah, so far as we can tell, was small, 
and it is not mentioned in the pre-Islamic fighting. Yet at the time 
of Muhammad's death its chief, Sa'd b. 'Ubadah, was the leading 
man not merely of the Khazraj but of the Ansar as a whole. As- 
Samhudi speaks of four subdivisions in separate localities; but 
these were presumably adjacent, and all were near the suq or 
marketplace of the Muslim city. Among the records are marriages 
with Maghalah, Salimah, Ba'l-Harith,Ba'l-Hubla, 'Abd al-Ash'hal, 
and Zafar. From an early time Muhammad seems to have been 
aware of the actual or potential importance of this clan, for, 
although only two members of it were present at the great con- 
vention at 'Aqabah, both of these became nuqabff or representa- 

1 Wellhausen, Medina, i8n.; IS, iii/2. 95; as-Samhudi, i. 141, &c. 


Salimah. The clan of Salimah along with Zurayq, Bayadah, and 
some small fragments belonged to B. Jusham b. al-Khazraj. It was 
a large clan, but in Ibn Sa'd its proportionate strength may seem 
greater than it really was, since its members were outstanding for 
their enthusiasm for Islam. It seems to have played little part in 
pre-Islamic politics, though it had some skirmishes with its neigh- 
bours on the east, 'Abd al-Ash'hal. Some families of Salimah were 
on friendly terms with some families of 'Abd al-Ash'hal, and these 
protected one another from their fellow clansmen. An unusually 
large percentage of the marriages recorded in Ibn Sa'd are within 
the clan, especially if nomadic confederates are included. There 
was also intermarriage with the other parts of B. Jusham, and with 
an-Najjar, Sa'idah, Harithah, 'Abd al-Ash'hal, Zafar, and 'Amr b. 
c Awf. As-Samhudi speaks of several distinct localities occupied by 
this clan, adjacent to one another, and near the foot of mount Sal'. 
The chief of Salimah was al-Jadd b. Qays, but Muhammad at 
some point caused him to be replaced by al-Bara' b. Ma'rur. 1 The 
latter, on pilgrimage to Mecca before the Hijrah, had refused to 
turn his back on the Ka'bah until told by Muhammad to face 
Jerusalem, and, appropriately, it was in the territory of Salimah 
that the change in qiblah was announced. This suggests that in 
some sections at least of Salimah there was no friendship for 
the Jews. 

Zurayq. Zurayq, Bayadah, and a small group called B. Hablb 
b. 'Abd Harithah had once been friendly with one another, but 
a quarrel developed and Hablb sided with Bayadah against Zurayq. 
Zurayq was forced to evacuate its lands, perhaps more than once, 
and these were occupied by Hablb. Some of Zurayq eventually 
emigrated to Syria. The matter was complicated by the fact that one 
family of Zurayq remained among Bayadah for a time, though at 
last it decided that life was better among its own clan; and one 
family of Hablb, becoming involved in a blood-feud with Zurayq, 
settled it by abandoning the hilfor alliance of Bayadah for that of the 
latter. Another blood-feud caused a small part of Jusham, B. Ghu- 
darah, to attach itself to Zurayq, because the other fragments of 
Jusham with which it was in collision received the support of 
Bayadah. The antagonism between these two clans links up with 
various points. As just seen, Ba'1-Hubla, who under ' Abdallah b. 

1 Cp. p. 234 below. For the spelling of the clan name see as-Suyutf, Lubb 
al-Lubdb t ed. P. J. Veth, Leiden, 1840-2, i. 138. 


Ubayy were opposed to Bayadah under 'Amr b. an-Nu'man, inter- 
marry with Zurayq and not with Bayadah, while their rivals the 
Qawaqilah do the opposite (unless the mother of Raff b. Malik 
is a real as well as an apparent exception). Harithah intermarry 
with Bayadah, and the enemies of both, f Abd al-Ash'hal, with 
Zurayq. According to a report in as-Samhudi Bayadah and Zurayq 
along with Zafar were the best of the Ansar in war, and this, if there 
is any substance in it, may have attracted small groups to them. 
Unfortunately, in the years round about the Hijrah, Zurayq did not 
produce any great leader. To it belonged Rafi' b. Malik, one of the 
twelve nuqabff or representatives appointed at al-'Aqabah, but he 
was not outstanding. 

Bayadah. Something has already been said about the ruthlessly 
aggressive policy of Bayadah under 'Amr b. #n-Nu'man. With his 
death at Bu'ath expansion stopped. One gets the impression that 
alliances were gladly made with aggrieved parties, such as Habib 
and Harithah, in order to have grounds for aggression. The small- 
ness of the number of men and women of the clan given notices by 
Ibn Sa' d may be due to great losses at Bu'ath or to lack of enthusiasm 
for Islam; or perhaps the clan had never been large. Its possession 
of nineteen utums or strongholds, however, according to as-Sam- 
hudi, indicates considerable military strength. Its territory, along 
with the territory of confederates like Habib and the other frag- 
ments of Jusham, formed a solid block, and this also made for 
strength. Apart from the other sections of Jusham and the clans 
just mentioned, Bayadah, like Zurayq, intermarried with B'al- 
Harith, an-Najjar, Sa'idah, and 'Amr b. 'Awf. 

It is convenient at this point to introduce the following table, 
although the figures in it refer to Muslims. But, except where a 
clan had a special reason for tending to accept or to reject Islam, 
we may, in default of better evidence, take these figures as a rough 
guide to the relative strength of the clans. For this purpose the last 
column is perhaps best, namely, the number of women to whom, 
as having sworn allegiance to Muhammad, a notice is given in 
Ibn Sa'd's eighth volume. In that column seven of those whom 
Ibn Sa'd classifies as Harithah have been transferred to 'Abd al- 
Ash'hal, since they appear to belong to the part of their clan which 
had become confederates of the latter. 



lims 1 

bahi 2 

bah2 3 



Uhud 6 


'Abd al-Ash'hal . 









. . 

. . 



liarithah . 




'Amr b. 'Awf 







Aws Manat (Khatmah) 

. . 


an-Najjar . 








al-IJarith . 

. . 






Ba'1-Hubla and 

al-Qawaqilah . 


































(c) Forces and tendencies in Medinan society 

This survey of the Medinan clans is precarious in that we do 
not know how complete our information is. If a serious gap has 
escaped notice or not been properly appreciated, a false and mis- 
leading emphasis may have been given at that point. Despite this 
possibility the basis of factual information is sufficiently wide to 
give a reliable general picture of the social forces and tendencies 
present in Meccan society. 

There are some instances of the tendency, constantly found in 
nomadic society, for large groups to disintegrate. One instance 
would be the quarrel of Zurayq with Hablb and then with Bayadah, 
another the separation of Silm from Waqif, and a third, as seems 
likely, the splitting up of the old clan of Salim. Under agricultural 
conditions the groups that attempt to live independently are 
smaller than would be the case in the desert. 

The main tendency, however, is a contrary one towards the 
formation of larger groups. In a society such as that of Medina, 
where there were numerous small groups in close contact with one 
another, it was always possible, if two had a quarrel, to appeal to 
third parties for help; and ambitious families and clans were 

1 IH, 287; cf. Caetani, Ann. i, p. 314. 

2 IH, 288-9; cf. Caetani, 1. c. 

3 IH, 305-12; cf. Caetani, ibid. 321-2. 

4 IH, 297-8; cf. Caetani, ibid. 319. 

5 IS, iii/2; cf. IH, 495-506; WW, 86-90; Caetani, ibid. 497-510. 

6 WW, 138-41; cf. IH, 607-10; Caetani, ibid. 563-4. 


usually ready to respond to such appeals. Normally, however, they 
had some definite grounds for their interference in other people's 
affairs. To a pre-Islamic Arab the most obvious of such grounds 
was kinship, even though kinsmen tended to have bitter quarrels 
with one another; and the genealogical theory of the two great 
tribes of the Aws and the Khazraj seems to have been worked out, 
whether with or without a genuine basis of fact, in order to justify 
and unify the two alliances which were splitting Medina. In the 
course of the survey there have been several examples of how 
genealogy seems to have been invoked to strengthen the ties which 
held together several small groups. It is possible that what was 
originally known to be a fiction for example, that Za'ura' was 
a 'brother' of f Abd al-Ash'hal in course of time came to be 
accepted as a genealogical fact. Allegations were also made about 
the eponymous ancestors in order to justify the existing relation- 
ship of the clans ; the father of Zurayq and Bayadah was said to 
have entrusted the former to the latter, and Zurayq was said on his 
death-bed to have given charge of his sons to Habib, who treated 
them harshly. In general, then, we have the curious position that 
kinship, though it was proving unable to prevent fratricidal strife, 
was in certain ways being developed as a principle of unity. 

The pre-Islamic Arabs were also familar with various forms of 
contractual agreement which bound men together. Chief among 
these were the mutual alliance (hilf y tahaluf) between .groups and 
individuals, by which they became confederates (hulafa?) of one 
another, and the jiwdr or temporary protection of a 'neighbour' 
(jar) in the Old Testament phrase, the 'sojourner within thy 
gates'. This method was employed in Medina. Thus we are told 
that B. al-Mu f alla broke off their hilf with Bayadah and formed 
one with Zurayq, while B. Ghudarah made a hilf with 'Amr b. 'Awf 
but then quarrelled with them and came to Zurayq. When, too, 
it is said that Za'ura' was 'brother' of 'Abd al-Ash'hal, I take this 
to mean that he had some such status as that of 'neighbour' and 
was allowed to intermarry. At the same time, it is possible that the 
later genealogists to whom we owe our information, when they 
were unable to link a man patrilineally with the clan to which they 
knew he belonged, assumed he was a confederate, whereas he may 
have been linked to the clan in the female line at some point. State- 
ments about confederates should therefore be treated with a certain 


In the growth of larger units the influence of physical or geo- 
graphical neighbourhood was important. For defensive purposes 
it became usual in Medina to employ the u\um or stronghold, and 
in order to build and maintain a stronghold a certain minimum 
number of persons was presumably required. Groups smaller than 
this minimum would therefore be compelled to join with other 
small groups, and so we find units like the 'people of Ratij' which 
was an amalgamation of fragments joined together for mutual 
defence, and becoming in course of time also linked by blood. In 
larger units, also, locality was important. Each of the main clans 
was in a sense a minute state, for it was an independent political 
entity. Within the territory of this state there was a measure of 
security, since to shed the blood of one's fellow clansman was an 
unpardonable offence.* Outside the territory of one's clan and its 
confederates there was little security and, in the period of 'cold 
war' after Bu'ath, positive danger. It is noteworthy that most of the 
cases recorded of intermarriage between clans are between adjacent 
clans. A man might venture a little way into the territory of another 
clan where he knew he had some friends; but to go right across 
another clan's lands to those of a third was a risky matter. 

In the development of larger units the personality of the leader 
played a great part. We are more likely to understand the history 
of the decade or two before the Hijrah if we concentrate not on the 
supposed hostility between the Aws and the Khazraj, but on 
the relations between the individual leaders who commanded in the 
main battles. These leaders must have seen that the existing state 
of affairs practically a war of all against all was intolerable, and 
that there were opportunities for a strong man to gain control over 
a large section of Medina, perhaps even over the whole. This was 
the issue at stake at Bu'ath. Had 'Amr b. an-Nu'man won there, 
no one in Medina would have been able to stand up to him. He 
had proved himself, however, to be little better than the leader of 
a robber band, ready, as far as possible, to meet the claims of the 
members of the band, but ruthless and unprincipled in his conduct 
towards those outside his band. He could promise his followers 
the rich lands of the Jewish clans of Qurayzah and an-Nadir 
before so far as we can tell any casus belli had appeared; and he 
could kill the Jewish hostages for a dubious reason. Such acts show 
that he was not sufficient of a statesman to look beyond his immedi- 
ate advantage and, since the unification of Medina was almost 


inevitable, to consider on what principles the would-be ruler of the 
whole must act. 'Amr's lack of principle was sowing the seeds of 
future conflicts and may well have lost him the support of men like 
'Abdallah b. Ubayy, who would feel that under such a leader there 
was little security. 

One wonders whether Ibn Ubayy had a wider vision. All we 
know is that he disagreed with c Amr b. an-Nu'man on the question 
of the Jewish hostages, since he set free those in his own hands, 
and that he remained neutral at Bu'ath. He may simply have been 
afraid of 'Amr; but it is probable that he realized the need for 
a single ruler in Medina (since his supporters are said to have been 
preparing to crown him when Muhammad arrived), and saw that 
this ruler must not lightly cause discontent in any section of the 
community, but must attempt to treat all ^parties fairly. Perhaps 
through his Jewish friends he had been influenced by Old Testa- 
ment ideals of social justice. 

Al-Hudayr b. Simak, the other commander at Bu'ath, may not 
have been much better than f Amr, since he had driven out Harithah 
from its lands. Yet there is nothing to make one suppose that he 
equalled ' Amr in ruthlessness, and the heterogeneous character of 
his supporters suggests that he was fair in his dealings with them. 
The degree of unity between diverse elements attained in the clan 
of 'Abd al-Ash'hal provided a pattern and was a good augury for 
the unification of Medina as a whole. 

Finally, it should be noticed that the clans which, according to 
the incomplete figures just quoted, were strongest numerically did 
not produce the strongest leaders. Ba'1-Hubla and Bayadah, the 
clans of Ibn Ubayy and 'Amr, were not large, and those that were 
large like an-Najjar and Salimah did not produce a great leader; 
even as As'ad b. Zurarah of an-Najjar was not on the same level as 
those mentioned. Al-Hudayr b. Simak was perhaps in a different 
position, since 'Abd al-Ash'hal was fairly numerous, but his rela- 
tion to it is obscure. This is a curious point and attention will have 
to be paid to it later. 


The three lists of names given by Ibn Ishaq, from which are 
derived the figures in the above table under the headings 'First 
Muslims 1 , "Aqabah i', and "Aqabah 2', may be taken to reflect 
three stages in the conversion of the Medinans to Islam, or, if one 


likes, in their conversion to the policy of bringing Muhammad to 
Medina. 1 When these figures are examined from the standpoint 
of clan relationships, some interesting facts come to light. The 
original approach for the alleged previous contacts and conver- 
sions did not lead to anything was made by men of an-Najjar, 
Zurayq, and Salimah, the foremost probably being As'ad b. 
Zurarah of an-Najjar. For the next stage, which is known as the 
first meeting of al-'Aqabah, these were joined by men from al- 
Qawaqilah, 'Abd al-Ash'hal, and 'Amr b. 'Awf. The second or 
great meeting at al-'Aqabah was attended by men from all the clans 
of the Aws and the Khazraj with the exception of Aws Manat (since, 
for this purpose, Zafar may be regarded as one with 'Abd al- 
Ash'hal). The noteworthy features are that, whereas the represen- 
tation of most clans Seems to be roughly in accordance with their 
strength, Salimah has proportionately three or four times as many 
representatives, and that 'Abd al-Ash'hal and Ba'1-Hubla, the clans 
of Sa'd b. Mu'adh, and 'Abdallah b. Ubayy, seem to be under- 
represented. Sa'd had become a Muslim prior to this, and so, 
presumably, had Ibn Ubayy, though in view of his later record we 
are told little about him. These two were the chief remaining 
leaders from the days before Bu'ath, and, though both became 
Muslims, neither deigned to be present at al-'Aqabah. It is to be 
noted that Ibn Ubayy and his like, who later came to be known 
as the Hypocrites, were for the first year or two after the Hjjrah 
just as much Muslims in outward conduct as any one else. In this 
Ibn Ubayy is contrasted with Abu Qays b. al-Aslat of Wa'il who 
did not become a Muslim, and with Abu 'Amir ar-Rahib of 'Amr 
b. 'Awf who retired to Mecca. 

These facts and figures give an idea of the groups which sup- 
ported Muhammad most vigorously in the early days of Islam in 
Medina. In particular the part played by Salimah is remarkable. 
Is it possible to discover why these clans and not the others were 
first attracted to Islam/t)ne feature common to nearly all the men 
in the first two lists is that they came from clans which had not 
produced great leaders themselves but which had suffered from 
warlike leaders belonging to other clans!\ An-Najjar and Salimah 
had been involved in fighting, but no military commanders from 
them are named. The sufferings of Zurayq have been mentioned. 
Al-Qawaqilah had produced the great leader, Malik b. al-'Ajlan, 

1 Ml Mecca, 144-7. 


but that was several generations earlier, and now they seemed to 
be weaker than Ba'1-Hubla and to be making friends with the 
latter's enemies. The clans of 'Amr b. 'Awf and 'Abd al-Ash'hal 
are in a diiferent position, but the two men at the first 'Aqabah 
seem to have been confederates and so probably did not represent 
the feeling of the leading groups within each clan. 

It is easy to see how clans and sub-clans in this position would 
be attracted by the prospect of an outsider coming to hold the 
balance in the affairs of Medina. They may have felt that unifica- 
tion was bound to come sooner or later, but have disliked being 
under a ruler from Ba'1-Hubla or Bayadah/ A member of any of 
the Medinan clans would already have his friends and his enemies 
among the other clans, and was unlikely to be fair to all. The fact 
that Muhammad was an outsider and the tfust his character in- 
spired in them gave a promise that he would be more satisfactory. 
Moreover, under Muhammad unification would be effected with- 
out first having a disastrous civil way Since Salimah showed anti- 
Jewish tendencies, one may also wonder whether they were afraid 
that Ibn Ubayy, who was friendly with the Jewish clans, might 
rely on them to a great extent and so allow them to regain their 
influence in Medina. It is tempting to suppose that the clans 
represented at the first 'Aqabah were proletarian in constitution 
and the remainder aristocratic, but there is not sufficient evidence 
for such a view. 

It remains to account for the conduct of the leaders. But for the 
conversion of Sa'd b. Mu'adh the course of the Islamic community 
would not have been so successful as in fact it was. His relation to 
al-Hudayr b. Simak, the victor of Bu'ath, is not clear, but he seems 
to have succeeded to some of his power, and, according to our 
sources at least, al-Hudayr's son Usayd worked harmoniously with 
Sa'd as his second in command. Yet this duality in the leadership 
of f Abd al-Ash'hal may have been felt by Sa'd as a weakness, and 
it is perhaps significant that even in Ibn Ishaq's pleasantly romanti- 
cized story of his conversion, it is made to follow soon after that 
of Usayd. This factor, however, was by no means the only one. 
Sa'd must be credited at the least with some awareness of what 
this new movement was going to mean in Medina, and at the most 
with a genuine belief in the message proclaimed by Muhammad. 
He was a kinsman of As'ad b. Zurarah and was friendly with 'Amr 
b. al-Jamuh of Salimah, so that he had good opportunities of 


estimating the hold Islam had gained on members of these clans, 
and of hearing the case for Islam presented by men towards whom 
he was sympathetically disposed. When he decided to become 
a Muslim, he entered Islam whole-heartedly; he had no hesitation 
in going on the expedition to Badr, from which Ibn Ubayy and 
even Usayd b. al-Hudayr stayed away (though the latter subse- 
quently apologized to Muhammad). As Sa'd b. 'Ubadah was also 
absent from Badr, allegedly suffering from snake-bite, Sa'd b. 
Mu'adh was the leading man among the Ansar there present and 
contributed much to the Muslim victory. 

The attitude of Ibn_JJbajry has to be surmised from the scant 
accounts of his behaviour. He must have realized that the move- 
ment towards Islam had become so strong that it could not be 
checked, and that to withstand it would simply cause him to lose 
influence/He probably also hoped to rule all Medina, and may 
even have seen in Islam a means towards his end; if Islam pro- 
vided a religious basis for unity, he could exercise the political 
control. Had the Jews been converted to Islam, this dream might 
have come true in party But the Jews rejected Islam, Muhammad 
proved to be an expert in handling political affairs, and Ibn Ubayy 
showed too little fervour for the cause he had nominally espoused 
to have a position of importance within the religious movement. 
This is admittedly conjectural, but some such line of thought is 
required to explain why Ibn Ubayy became a Muslim, and then, 
without outwardly ceasing to be a Muslim, became an opponent 
of Islam. 

It may well be, as Ibn Ishaq says, that it was the conversion of 
Sa'd b. Mu'adh which led to the general acceptance of Islam by 
the Medinans. Study of the available evidence gives the impression 
that, apart from the exceptions about to be mentioned, the accep- 
tance was really general, and that all the leading men and women 
in all the clans became at least nominally Muslims. We hear of no 
person of even a moderate degree of importance who became a 
Muslim only after, say, the victory of Badr or the breaking of the 
siege. In other words we have justification for assuming that there 
is no important person of whom we are uninformed, though it is 
likely that there are many mediocrities of whom we know nothing. 

The clans from whom before the Hijrah there were no converts 
of standing were those of Aws Manat, namely, Umayyah b. Zayd 
(of Murrah b. Malik b. al-Aws), Khatmah, Wa'il, and Waqif. The 

5783 N 


Jewish clans also kept aloof from Islam, though in the course of 
time there were a few converts from them. None of the Arab clans 
mentioned seems to have been strong, though there was among 
them in Abu Qays b. al-Aslat of Wa'il a leader who at one time 
had been in the first rank. The sources do not give any reason for 
the refusal of these clans to join the new movement. Abu Qays is 
said to have thought of becoming a Muslim, but to have died before 
he put his thought into effect ; such thoughts without actions, how- 
ever, make one suspect an attempt to save the face of the clan, for 
the one solid fact which is not denied is that Abu Qays did not 
become a Muslim. The most probable explanation of this lack of 
response to Muhammad is that these clans were closely linked 
with the Jews. Their lands were apparently not in a solid block, 
but mixed among Jewish lands. Their position was thus weak, and 
it is understandable that they were not ready, without further 
observation at close quarters, to commit themselves irrevocably to 
a movement that was looked on with disfavour by their Jewish 

It is convenient at this point to narrate the subsequent history of 
this 'pagan opposition', since it never was of prime importance 
in the affairs of Medina. Abu Qays died before Badr, and the other 
leading men also held aloof from Muhammad, though there were 
some converts among the rank and file, presumably some younger 
men. Those who remained pagans were bitter about the advance 
of Islam. In particular, ' Asma' bint Marwan (of Umayyah b. Zayd 
of Aws Manat), the wife of a man of Khatmah, composed verses 
taunting and insulting some of the Muslims. 1 If those quoted by 
Ibn Ishaq are genuine, the chief point was that the persons 
addressed were dishonouring themselves by submitting to a 
stranger not of their blood. Shortly after Badr (according to the 
most probable version) a man of Khatmah called 'Umayr b. 'Adi 
(or 'Udayy) went to the house of 'Asma' by night and killed her. 
Muhammad did not disapprove, no one dared take vengeance on 
'Umayr, and many of the clan (and perhaps of the rest of Aws 
Manat) now professed Islam openly; some of these are said to 
have been secret believers previously. The assassination of Abu 
'Afak of 'Amr b. 'Awf about the same time 2 by a man of his clan 
had similar motives and probably similar effects, since some 

1 IH, 995-6; WW, 90-91. 

2 IH, 994-5; WW, 91-92. WW's dating is to be preferred. 


sections of 'Amr b. e Awf were close to Aws Manat both in outlook 
and in physical situation. Abu e Afak had taunted his hearers with 
allowing an outsider to control their affairs, a man who confused 
right and wrong and who aimed at kingship. After these events we 
may assume that there was little overt opposition to Muhammad 
among the pagans. If any still refused to become Muslims, they 
must have been relying on Jewish support, and can hardly have 
continued in their refusal after the expulsion of the chief Jewish 
clans. By the time of the campaign of Hunayn we find among the 
Muslim forces contingents from Waqif, Khatmah, and Umayyah. 1 
Thus, because of the bankruptcy of paganism confronted with the 
gloomy situation in Medina, the pagan opposition with its appeal 
to old ideas of honour and blood-relationship gradually died out. 
The sons of the pagans became Muslims, the idols of the clans 
were destroyed, 2 and Aws Manat came to be known as Aws Allah. 

The case of Abu ' Amir ar-Rahib is mysterious, but may be dealt 
with here since he was neither Jew nor Muslim. 3 His name was 
'Abd 'Amr b. Sayfi, and he belonged to the sub-clan Dubay'ah 
of 'Amr b. 'Awf. For many years before the Hijrah he had been 
a monotheist, and by his ascetic practices had gained the nickname 
of ar-Rahib, the monk, although his asceticism did not include 
celibacy. When Muhammad came to Medina, Abu 'Amir, rather 
than submit to him like his maternal cousin Ibn Ubayy, migrated 
to Mecca. With fifteen (or perhaps fifty) Medinan followers he 
fought against the Muslims at Uhud. He seems to have been at 
Khaybar for a time. On the fall of Mecca he retired to at-Ta'if, 
and when that also submitted took refuge in Syria. Of his outlook 
we can know only what is to be inferred from his actions. He is 
important, however, as a sign that Muhammad's claim to be a 
prophet was a stumbling-block to some who on general grounds 
might have been expected to welcome the new religion ; and pre- 
sumably this was so because they were aware of the political 
implications of the claim and disliked them. 

Thus, with the exception of Abu Qays and Abu 'Amir and their 
meagre following, Muhammad, when he went to Medina, had the 
support for one reason or another of all the most influential men 
among the Arabs; and, apart from Sa'd b. Mu'adh and Ibn Ubayy, 

1 WW, 358 and n. 2 IS, iv/a. 94-95, 90. 

3 IH, 411-12, 561; WK, 205-6; WW, 103, 190, &c.; IS, viii. 251-2 


they all came to the great convention of al-'Aqabah. It is worth 
giving the names of the twelve nuqabff or representatives who were 
appointed there, for they were leading men in their clans and in 
the new Islamic community as a whole. 

Usayd b. al-Hudayr .... 'Abd al-Ash'hal. 
(Abu '1-Haytham b. at-Tayyihan) . ,, 

Sa'd b. Khaythamah .... 'Amr b. 'Awf. 

Rifa'ah b. c Abd al-Mundhir . . 'Amr b. r Awf. 

Sa'd b. ar-Rabi' .... Ba'1-Harith. 
'Abdallah b. Rawahah ... ,, 

Sa'd b. 'Ubadah . Sa'idah. 
al-Mundhir b. 'Amr . . . . ,, 

al-Bara' b. Ma'rur .... Salimah. 

'Abdallah b. 'Amr b. Haram . . " 

'Ubadah b. as-Samit .... al-Qawaqilah. 

Rafi' b. Malik Zurayq. 

As'ad b. Zurarah .... an-Najjar. 


The opponents of Muhammad among those who had formally 
professed Islam are commonly spoken of as the mundfiqun or 
Hypocrites, and the usage has Qur'anic sanction. A more useful 
term in the present connexion, however, is the 'Muslim oppo- 
sition' since this name distinguishes the object of study from the 
pagan opposition (just mentioned) and the Jewish opposition (to 
be dealt with later), and does not restrict the historian to those 
persons branded as Hypocrites. There were occasional disagree- 
ments with Muhammad's policy even among those Muslims loyal 
to him, but the sources tend to minimize the disagreements within 
the community and to suggest that it was more united than in fact 
it was. Only in the case of those stigmatized as Hypocrites are we 
given accounts of what they said and did against Muhammad, and 
even these are meagre. There is consequently a scarcity of informa- 
tion about the internal politics of Medina, and at many points we 
have to rely on conjectures and probabilities. 

(a) The first five years 

The setting out of the expedition which humbled the Meccans 
at Badr was the first occasion on which the Ansar were faced with 


an important decision. Were they to respond to Muhammad's 
summons? One who did not respond was Usayd b. al-Hudayr; but 
on the return of the victorious army he apologized to Muhammad, 
saying that he thought it was only a raid for booty, and that had he 
known there was to have been fighting he would certainly have 
been present. This conduct could be explained easily as a reaction 
to Muhammad's success. We might suppose that Usayd, learning 
that Sa'd b. Mu'adh (his rival for leadership within the clan of 
f Abd al-Ash'hal) was high in Muhammad's counsels, dallied with 
the idea of gaining an advantage by not identifying himself with 
what probably seemed to be a losing cause; when it was made clear 
that the cause was not a losing one, he hurriedly dropped this idea 
and resigned himself to second place within the clan on the side of 
Muhammad. 1 

This explanation, even if mainly sound, probably has a false 
emphasis, laying too much stress on the personal rivalry. Up to 
this time, some eighteen months after the Hijrah, Muhammad had 
apparently accomplished nothing of moment. Others also stayed 
away Ibn Ubayy, and even Sa'd b. 'Ubadah; the latter indeed is 
said to have been suffering from snake-bite, but that may be merely 
his excuse. 2 If there was a movement away from Muhammad, his 
notable victory and his gentle handling of the truants put a stop 
to it. Usayd and Sa'd b. 'Ubadah continued to stand high in his 
favour, and even Ibn Ubayy did not refuse outright to help at Uhud. 
The man who gave the lead in loyal devotion to Muhammad was 
Sa'd b. Mu'adh, and he continued to be foremost of the Ansar as a 
wholeuntil his death, when Sa'd b. 'Ubadah (Sa'idah) took his place. 
The leader of the Khazraj at Badr was al-Hubab b. al-Mundhir 
(Salimah), but he was not specially prominent in later events. 

If there was an incipient movement away from Muhammad 
among the Muslims, probably encouraged by the Jews, it is con- 
ceivable that the attack on the Qaynuqa' was intended by Muham- 
mad not merely to weaken the Jews but to reward his supporters 
and to teach a lesson to the lukewarm Muslims like Ibn Ubayy. 
It is noteworthy that Qaynuqa' were confederates of Ibn Ubayy, 
who had fought by his side on several occasions before the Hijrah. 
Among those prominent in the attack on Qaynuqa' were Sa'd b. 
Mu'adh, who must be reckoned a rival of Ibn Ubayy, and 'Uba- 
dah b. as-Samit of al-Qawaqilah, which, as has been seen, was 
1 IH, 428; WW, 37 f-> 72. * WW, 66. 


apparently jealous of Ibn Ubayy and Ba'1-Hubla. 'Ubadah had also 
been a confederate of the Jews (doubtless because he and Ibn Ubayy 
were both members of the old B. Salim), but, instead of pleading 
for them like Ibn Ubayy, he declared that he renounced his relation- 
ship to them. 1 

Ibn Ubayy appears to have used the argument to Muhammad 
that the Meccans were likely to march against Medina to exact 
revenge, and that in such a case the support of the 300 armed men 
of Qaynuqa' would be an asset, and that therefore they should not 
be expelled. According to a tradition which may be accepted, the 
following Qur'anic passage refers to this occasion : 

ye who have believed, do not take Jews and Christians as friends 
(or patrons); they are friends to each other; whoever of you makes 
friends of them is one of them ; verily God dotfi not guide the wrong- 
doing people. Yet one sees those in whose hearts is disease hastening 
(sc. to speak to Muhammad) about them, saying: 'We fear a turn of 
fortune may befall us/ But possibly God will bring the Issue (or final 
deliverance) or some affair from Himself (sc. direct intervention), and 
they will become remorseful on account of what they concealed within 
themselves. . . . 2 

The following verses, which exhort the believers to 'take as friend 
(or patron) God and His messenger and those who have believed', 
are traditionally connected with the action of 'Ubadah in renounc- 
ing his league with non-Muslims. 

This passage from the Qur'an tends to prove what general con- 
siderations made a probability, namely, that this incipient oppo- 
sition to Muhammad among Medinan Muslims was strongest 
among those friendly with the Jews. Other passages corroborate. 
The phrase 'those in whose hearts is disease' is the Qur'anic term 
at this period for the Muslim opposition. This group is accused 
of criticizing the Qur'an, especially some of the more recently 
revealed passages. 3 In particular they made difficulties over the 
matter of abrogation. 4 To begin with, their main effort seems to 
have been to weaken Muhammad's position by verbal arguments. 5 
When it became clear, however, that Muhammad's policy was to 
provoke the Meccans by raids on their caravans which might, like 
that to Nakhlah, involve bloodshed, they became seriously alarmed. 

1 IH, 545-7; WW, 92-93. 2 Q. 5- 51 f./S7 f- 

3 Q. 9. 124-7/125-8 E; cf. 74. 31 E-K 

4 Q. 22. 52/51-54/53 B. 5 Q. 2. 8/7-15/14 FG. 


Some time before Badr the Muslims had received by revelation 
the command : 'when ye meet those who have disbelieved (let there 
be) slaughter . . . until war lays down its burdens'. l We may surmise 
that the Ansar had not been enthusiastic about the matter until 
this revelation came, for it is probably in this connexion that the 
Qur'an reports them as saying: 'Why has not a surah been sent 
down?' 2 The loyal Muslims now accepted the policy of provoking 
Quraysh, but the opposition became more alarmed; 'when a clearly 
formulated surah is sent down and fighting is mentioned in it, thou 
seest those in whose hearts is disease looking at thee with the look 
of one already faint in death'. 3 About the same time (or perhaps 
a little later) there was a movement to avoid bringing disputes to 
Muhammad for settlement. 4 

The Qur'an and the biographical sources thus give complemen- 
tary pictures of the discontent in Medina with Muhammad's policy. 
His success at Badr and against Qaynuqa', however, and his firm 
but gentle handling of the opposition prevented any serious 
attempt to leave his camp for that of the Meccans, and when 
Quraysh advanced against Medina in the campaign of Uhud, the 
Islamic community was intact, if not altogether united. Ibn Ubayy 
had an honoured place in the discussions of strategy. He sup- 
ported Muhammad's original suggestion, that they should remain 
in the strongly fortified central settlements ; but it may well be that 
Muhammad made the suggestion because he knew that such a 
policy of playing for safety was most likely to be acceptable to 
men like Ibn Ubayy. The young men, eager for battle, protested, 
and found some responsible men to support them on the grounds 
that to do nothing while their crops were being ruined would cause 
a serious loss of prestige. This group opposed to Ibn Ubayy in- 
cluded Hamzah (Muhammad's uncle), Sa'd b. 'Ubadah (Sa'idah), 
an-Nu'manb. Malik (al-Qawaqilah), lyasb. Aws ('Abd al-Ash'hal), 
and Khaythamah and Anas b. Qatadah ( c Amr b. c Awf). This list 
of names is not large, and it is possible that many Muslims favoured 
the first course. Sa'd b. Mu'adh and Usayd b. al-Hudayr may have 
done so, for, when Muhammad decided on the second course, they 
seemed to think that he had been unduly influenced by the pressure 
of the second group, and proposed that the decision should be 
reviewed and left entirely in his hands. 

1 Q. 47. 4 EF. * Q. 47. 20/22 EF; cf. 35/37. 

3 Ibid. 20/22. 4 Q. 24, 47/46~52/5i G- 


What subsequently happened has already been narrated. Mu- 
hammad adhered to the decision already taken, and Ibn Ubayy, 
after marching out part of the way, retired with his party to 
their strongholds. With reference to this and to Ibn Ubayy's 
remarks to his wounded son after the battle, the Qur'an says: 

What befell you on the day when the two hosts met was by the per- 
mission of God, and in order that He might know the believers and in 
order that He might know those who played the hypocrite; they were 
asked to come and fight in the way of God, or to defend (themselves), 
but they said : 'If we knew aught of fighting (sc. with a chance of success ; 
or else, if we thought there would actually be fighting), we would follow 
you.' They were that day nearer to unbelief than to belief, saying with 
their mouths what was not in their hearts ; but God knoweth what they 
conceal those who, having stayed behind, say regarding their brethren : 
'If they had obeyed us, they would not have been killed.' 1 

The word here translated 'played the hypocrite' properly means 
'crept to their holes' like moles or mice. In this passage it is prob- 
ably used for the first time with regard to Ibn Ubayy and his party, 
and used in its literal sense ; later, of course, the participle mund- 
fiqun became the regular description of the 'Muslim opposition* 
and is commonly translated Hypocrites; but the derived sense is 
probably due to this Qur'anic passage, and we might perhaps 
convey more of the original feel of the word by speaking of the 
Creepers or the Moles. 2 

It is noteworthy that the Qur'an here speaks only of the cowar- 
dice of the Hypocrites, and does not accuse them of disobedience. 
From this it is to be inferred that Ibn Ubayy was within his rights 
in acting as he did, and did not formally break his league with 
Muhammad (which was presumably in the terms of the Constitu- 
tion). 3 Previously he had gone out of his way to be condescending 
towards Muhammad in public ; now he made it clear that he was 
not a wholehearted supporter of Muhammad like Sa f d b. Mu'adh, 
but insisted on being regarded as at least an equal. Since he was 
still nominally a Muslim and had committed no punishable offence, 
Muhammad, who was occupied with promoting public security, 
could take no violent measures against him, much as his followers 
desired this; but when he rose in the mosque after the Friday 

1 Q. 3. 166/160-168/162 G; cf. WW, 145. 

* But cf. A. Jeffery, The Foreign Vocabulary of the Qur'dn, Baroda, 1938, 
s.v. ; Horovitz, Koranische Untersuchungen, 64, with further references. 
3 Cf. p. 221 below. 


Worship, and tried to make his usual condescending speech, he was 
roughly handled by men of Malik b. an-Najjar and al-Qawaqilah. 1 
Up to this point Ibn Ubayy and his friends had been grumbling 
and criticizing Muhammad and his revelations, but their position 
had essentially been that of 'sitting on the fence'. It is presumably 
they who are described in these words of the Qur'an: 'Those who 
have believed and then disbelieved, then believed and then dis- 
believed again, . . . vacillating between (this and) that, are neither 
one thing nor the other.' 2 His treatment after Uhud, however, 
seems to have infuriated him, and for the next two years he was 
seeking opportunities of injuring or even destroying Muhammad. 
Five months after Uhud the Muslims sent an ultimatum to the 
Jewish clan of an-Nadir. Ibn Ubayy and some fellow clansmen 
did all they could to* persuade the Jews to resist, even promising 
military support. Some of the Jews found this to their liking, but the 
cooler and wiser heads were aware of the emptiness of the pro- 
mises, and an-Nadir soon submitted. Their departure was a further 
defeat for Ibn Ubayy. His actions are described in Surat al-Hashr : 

Hast thou not seen those who have played the hypocrite saying to 
their brethren the People of the Book who have disbelieved: * Surely, if 
ye are expelled, we shall go out with you, we shall never obey anyone 
in regard to you, and if ye are attacked in war, we shall help you* ? God 
testifieth that they are lying. If they are expelled, they (the Hypocrites) 
will assuredly not go out with them, and if they are attacked in war, 
they will not help them, and if they help them, they will certainly 
turn their backs (in flight), and then they will not be helped. 3 

The next opportunity for action, so far as our records go, was 
over a year later during the expedition of Muraysi*. A quarrel 
between two men over a bucket of water rapidly developed into 
a fight between the Ansar and the Emigrants, and the results might 
have been serious had it not been as rapidly quelled by the loyal 
Muslims. Ibn Ubayy seems to have used the occasion to say to any 
who would listen that this man who came ostensibly to keep the 
peace was merely involving them in brawls; and he also seems to 
have muttered something about the stronger driving out the weaker 
when they returned to Medina. As Surat al-Munafiqin puts it, it is 
the Hypocrites 'who say: "If we return to the city, the highest in 
dignity in it will assuredly expel the most abased/ 1 though dignity 

1 IH, 591 f.; WW, 145. 2 Q. 4. 137/136, 142/141 F; cf. 141/140. 

3 Q. 59. ii f.; IH, 652 f.; WW, 162-5. 


belongs to God and His messenger and the believers'. Muhammad 
was informed of Ibn Ubayy's words, but refused to take any violent 
action, though Ibn Ubayy's son 'Abdallah, who was devoted to 
Muhammad, said he would himself kill his father if Muhammad 
wanted that done. Muhammad preferred to tire out the participants 
in the expedition by an exceptionally long march. 1 

Ibn Ubayy, however, had not learnt his lesson. Before the party 
arrived back in Medina, Muhammad's young wife 'A'ishah, who 
had accompanied them, was somehow left behind after the last 
halt, and eventually entered Medina after the others attended by 
a handsome young man. Tongues wagged, and Ibn Ubayy did 
what he could to magnify the scandal. In this he had some strange 
helpers, the son of a cousin of Abu Bakr, the poet Hassan b. Thabit 
and the sister of Muhammad's wife Zaynabf, each of whom must 
have been moved by some personal animosity against 'A'ishah or 
by sympathy for Ibn Ubayy or by dislike for the Emigrants. The 
scandal kept growing for weeks before matters came to a head. 
The question of fact was decided by Muhammad in favour of 
'A'ishah, since there was no solid evidence against her; and the 
incident is commonly referred to as 'the affair of the lie' ('ifk). 
The lesser scandal-mongers are said to have been flogged. With 
Ibn Ubayy Muhammad had a 'show-down'. He summoned a meet- 
ing of the leading men among the Ansar, and asked for permission 
to take punitive measures against one of them who was attacking 
his family; violence towards a man without the consent of his clan 
or tribe would lead to reprisals by the lex talionis. Muhammad's 
request was followed by an angry scene in which the Aws and the 
Khazraj nearly came to blows a state of affairs perhaps deliber- 
ately provoked to make them forget their common grievance 
against the Emigrants. It was not long before the quarrel was 
made up. 2 

Ibn Ubayy was probably not punished in any way (though some 
authorities said he was flogged). From this point onwards, however, 
there are no records of his taking any active steps against Muham- 
mad, and it may be assumed that he now realized that his following 
was so small that he could not hope to achieve anything. He was 
too old to become an enthusiastic Muslim, and he may sometimes 

1 Q. 63. 8; IH, 726-8; WW, 179-83. 

2 IH, 731-40; WW, 184-9; al-Bukharl, Maghdzl (64), 34; Nabia Abbott, 
Aishah the Beloved of Mohammed, Chicago, 1942, 29-38. 


have grumbled, 1 but he was sufficiently a Muslim at al-Hudaybiyah 
in 628/6 to refuse the privilege of making the pilgrimage, which 
Quraysh, while denying it to the other Muslims, offered to him as 
a special favour. 2 That he did not remain unreconciled to Muham- 
mad is further shown by his presence at al-Hudaybiyah and by 
the fact that Muhammad himself conducted his funeral rites. 3 

The weakness of the position of Ibn Ubayy was that it had no 
ideological basis. As one of the leaders of an-Nadlr is alleged to 
have put it, Ibn Ubayy did not know what he wanted; he was 
whole-heartedly committed neither to Islam nor to Judaism nor 
to the old religion of his people. 4 He was probably moved chiefly 
by personal ambition, and lacked the statesmanship to see all the 
vaster issues involved and the vision to propound a way of dealing 
with them that would* attract men. He must have seen the need for 
peace in Medina, but his attempts to meet it were along conserva- 
tive lines that were already discredited. His opposition to Muham- 
mad may be said to be due to a failure to move with the times; and 
it is significant that one source remarks that there was only one 
young man among the Hypocrites. 5 

Perhaps there were similar reasons for the refusal of some men 
of 'Amr b. 'Awf to help with the defence of Medina when it was 
besieged. 6 Such people might be nominally Muslims, but they 
evidently did not regard membership of the Islamic community 
as a primary fact in their lives. Another whose attitude was pre- 
sumably similar was al-Jadd b. Qays, whom Muhammad deposed 
from the leadership of the clan of Salimah. 7 The Qur'an charges 
the Hypocrites with unbelief at the time of the siege, and gives the 
impression that their attitude was more dangerous than is indi- 
cated by the narrative sources, which were not written down until 
long after the great triumphs of Islam. 

The Hypocrites and those in whose hearts is disease were saying: 
'God and His messenger have promised us nothing but illusion. 1 A 
party of them said: 'O people of Yathrib, there is no abiding place for 
you, so return*; and a part of them were asking leave of the prophet, 
saying: 'Our houses are a weak point' ; they were not a weak point, they 
were only wishing to flee. If an entrance had been made upon them from 
that side (sc. on which their houses stood), and they had been asked to 

1 Cf. WW, 247- 2 Ibid. 255- 3 IH, 927; WW, 414 f. 

4 WW, 162. 5 IH, 363- 6 WW, 194. 

7 IH, 309; IS, iii/2. 112; for his conduct cf. IH, 746, 894; WW, 248, 392; 
cf. p. 234 below. 


join in sedition, they would have joined in it and would have hesitated 
but slightly. Yet they had covenanted with God previously that they 
would not turn their backs. . . .' 

(b) The last five years 

The treachery of Abu Lubabah in connexion with Banu Quray- 
zah may be said to mark the transition to the second phase of the 
opposition in Medina, when its efforts were directed not against 
the Islamic community as such but against particular aspects of 
Muhammad's policy. Unfortunately the affair of Abu Lubabah is 
obscure. The following is Ibn Ishaq's version. 

Then they (B. Qurayzah) sent to the Messenger of God (God bless 
and preserve him) the request, 'Send to us Abu Lubabah . . . that we 
may consult him about our course of action. 1 The Messenger of God 
(God bless and preserve him) sent him to them, and when they saw 
him . . . they said to him, 'O Abu Lubabah, do you think that we should 
surrender at the discretion of Muhammad?' He said, 'Yes,' and pointed 
with his hand to his neck, indicating that it would be slaughter. (Subse- 
quently) Abu Lubabah said, 'By God, my foot had not moved from the 
spot beforel realized that I had betrayed God and His messenger/ Then 
he departed. He did not go, however, to the Messenger of God (God bless 
and preserve him), but bound himself to one of the pillars in the mosque 
and said, 'I shall not leave my place here until God pardons me for what 
I have done/ and he swore to God, 'I shall never again go to Banu 
Qurayzah, and I shall never again go to a district in which I betrayed 
God and His messenger. . . . When news of this reached the Messenger 
of God (God bless and preserve him) he had been surprised at the 
delay he said, 'Had he come to me, I would have forgiven him; but 
since he has done as he has done, it is not for me to loose him from 
his place until God pardons him.' 2 

Abu Lubabah then remained bound to the pillar except during the 
times of prayer when his wife (or daughter) untied him. After six 
days Muhammad announced that God had pardoned him, and at 
Abu Lubabah's request himself untied him. 

The story as we have it must have been manipulated. The only 
obvious 'treachery' in it is the betrayal of Muhammad's intention 
of putting the men to death, for this might be supposed to have 
made Qurayzah less ready to surrender. Actually, however, it does 
not appear to have done so; and the offence is not commensurate 
with the punishment. Nor is the mystery explained by the further 
1 Q. 33- 12-15. 2 IH, 686 f.; cf. WW, 213-15. 


details in al-Waqidi of how Abu Lubabah tried to persuade the 
main body of Qurayzah to abandon Huyayy, who had been mainly 
responsible for the resistance to Muhammad. The explanation is 
probably to be looked for in some undertaking given by Abu 
Lubabah to stand faithful to his clan's alliance with Qurayzah. 
This would have led to a grave split in the Islamic community, 
had any attempt been made to punish Qurayzah. In this or some 
similar form the question at issue must have been the continuation 
of pre-Islamic relationships with non-Muslims. There is no sug- 
gestion that Abu Lubabah was other than a faithful member of 
the Islamic community ; he had no thought of leaving that com- 
munity, but on a certain matter of policy he differed from Muham- 
mad. This, then, is the characteristic of the Muslim opposition to 
Muhammad during the last few years of his life. It accepts the 
community as a fact but disagrees with particular lines of policy, 
usually for selfish reasons. 

The next case of this is the refusal of some of the nomadic tribes 
to take part in the expedition of al-Hudaybiyah in 628/6, doubtless 
because they failed to see any immediate gain to be derived from 
it. 1 Muhammad's grasp of events was so much wider than that of 
the majority of his followers that it must often have been difficult 
to bring them to accept his policies when these involved hardship. 
The crisis came with the expedition to Tabuk in 630/9. Arabia was 
now rushing to enter into alliance with Muhammad, and some of 
the worthy farmers of Medina thought it was time to have a little 
rest from their labours and enjoy their hard-won prosperity. Not 
so Muhammad. He realized that the internal peace of Arabia 
could only be maintained if its excess energies were directed out- 
wards. The expeditions to the north were thus of primary impor- 
tance in the creation of a stable Arab state. But this was a long-term 
policy whose advantages were not obvious to the multitude. In 
particular, some of the well-to-do men of Medina objected both 
to the discomfort of personal participation in an expedition and 
to the contributions (sadaqdt) they were expected to give. They 
kept their hands tight shut, and jeered at those who gave gener- 
ously. 2 When the summons came to march to Tabuk, a number of 
the Ansar remained at home. 3 

Three incidents connected with this expedition throw light on 

1 IH, 740; WW, 242. 2 Q. 9. 75/76-80/81. 

3 IH, 897, 907-13; WW, 393, 411-13- 


the state of feeling among some of the Muslims. There is said to 
have been a plot against Muhammad; something was to happen 
on a dangerous bit of road on a dark night, and it would have 
looked like an accident. 1 Then there was the 'mosque of dissension* 
(masjid ad-dirar). Just before the expedition set out Muhammad 
had been asked to honour by his presence a mosque at Quba' 
which some Muslims had built, but he postponed the matter till 
his return. On the journey, however, he somehow realized that an 
intrigue against himself was involved, and as soon as he returned 
to Medina he sent two men to destroy the mosque. The mosque- 
builders were of the clan of c Amr b. 'Awf and apparently supporters 
of Abu 'Amir ar-Rahib (who may even himself have been in 
Medina at this time), and the new mosque was to give them a con- 
venient meeting-place where they could hatch their plots without 
interruption. Abu Lubabah had made a gift for the mosque but 
was clear of the intrigues. 2 

About the same time the men who had stayed at home from the 
expedition to Tabuk were being cross-examined and their excuses 
scrutinized. Three who were not involved in the intrigue of the 
mosque-builders but had no good excuse were 'sent to Coventry 1 
for fifty days. The severity of the punishment shows the impor- 
tance of the matter, and several Qur'anic verses indicate that about 
this time those now called 'Hypocrites' were practically excluded 
from the community; they were to be treated roughly and threat- 
ened with Hell as apostates. 3 A little reflection makes it Clear that, 
if the Islamic community was to engage in expeditions into Syria 
which would involve the absence of most of its fighting men for 
long periods, it could not allow a body of dissidents to ensconce 
themselves in a suburb of Medina. Moreover, for the spiritual 
health of the community it was desirable that all its able-bodied 
men should share in the campaigns. Thus it would seem that about 
this time there was a definite change of policy towards the oppo- 
sition; but the Hypocrites who are now attacked and denounced 
are not identical with the previous ones, and may be entirely 
different. Ibn Ubayy was not at Tabuk, but his excuse (presumably 
ill-health) was apparently accepted, and he was not 'excommuni- 
cated', since Muhammad attended his funeral shortly afterwards. 

This series of events may well be regarded as the final crisis 

1 WW, 409. * IH, 906 f. ; WW, 410 f. 

3 Q. 9. 73/74 f-;cf. 66. 9. 


during Muhammad's lifetime in the internal politics of Medina. 
Medina was now prosperous, and some of the Ansar hoped for 
a lazy enjoyment of their prosperity. Muhammad, however, either 
persuaded them to accept his policy of continued expansion or 
showed them that his demands were not lightly to be rejected, 
since his will could be made effective by overwhelming force. In 
this way he established the Islamic community on foundations 
sufficiently solid to permit its expansion into an empire. 

There were other tensions in Medina, of course, notably those 
between Emigrants and Ansar 1 and between the Aws and the 
Khazraj, but the sources give little information about them during 
Muhammad's lifetime. It is sometimes only years afterwards that 
the cleavages become apparent. 

1 Cf. IH, 912; WW, 413; of the Emigrants only Talhah and az-Zubayr showed 
friendship to Ka'b b. Malik. 



THAT there were Jews in Medina when Muhammad went there 
is clear, 1 but how they came to be there and whether they 
were of Hebrew stock is not clear. Were they the descendants 
of fugitives from Palestine perhaps after the rising of Bar 
Kokhba? Were they mainly Arabs who had adopted the Jewish 
faith ? Such questions have been much discussed first by Muslim 
and then by Western scholars, but no general agreement has been 
reached. 2 The Jewish tribes had many custorqs identical with those 
of their pagan Arab neighbours and intermarried with them, 3 but 
they adhered firmly to the Jewish religion, or at least to a form of 
it, and maintained their distinct existence. 

When the Aws and the Khazraj came to Yathrib from the south, 
they found it dominated by Jews, though there were also a few 
Arabs in a subordinate position to the Jews. The dividing line 
between Arabs of this earlier stratum and Jews is confused. The 
Arabs were weaker than the Jews thirteen Arab strongholds 
(dtdm) to fifty-nine Jewish ones is one figure 4 and were in rela- 
tions ofjiwdr or hilf to them, that is, were protected by them, either 
as * neighbours* or as confederates. They probably intermarried, 
and marriage was presumably uxorilocal. 5 They may have adopted 
the Jewish religion. Not surprisingly, then, certain Arab clans are 
sometimes reckoned as Jewish clans; thus as-Samhudi's list of 
Jewish clans includes B. Marthad, B. Mu'awiyah, B. Jadhma', 
B. Naghisah, B. Za'ura', and B. Tha'labah, although the first of 

1 General sources for the section: as-Samhudi, 109-16, 152 ff. (= Wiisten- 
feld, Medina, 25-31); Ibn al-Athir, i. 400-20; Wellhausen, Medina, 7-15; H. 
Hirschfeld, 'Essai sur Thistoire des Juifs de Me*dine', Revue des fitudes Juives, 
vii. 167-93 1 x - IO ~3 i ; A. J. Wensinck, Mohammed en dejoden te Medina, Leiden, 
1928, 33-53 (part of this has been translated by G. H. Bousquet and G. W. Bous- 
quet-Mirandolle as 'L'lnfluence Juive sur les Origines du Culte Musulman', in 
Revue Africaine, xcviii (1954), 85-112). 

2 Cf. Wellhausen, I.e. ; Caetani, Ann. i. 383 ; C. C. Torrey, The Jewish Founda- 
tion of Islam, New York, 1933, ch. i ; D. G. Margoliouth, The Relations between 
Arabs and Israelites prior to the Rise of Islam, London, 1924, lecture 3. 

3 E.g. IH, 35 1 (parents of Ka'b b. al-Ashraf ) ; IS, viii. 236. 23 ff. (Umamah bint 
Bishr and Asad b. 'Ubayd). 4 As-Samhudi, 116 (= Wustenfeld, 31). 

5 Cf. p. 379 below; 


these is properly a part of the Arab tribe of Bali, the second a part 
of Sulaym, the third and fourth Arabs of the Yemen, and the last 
two Arabs of Ghassan. 1 

The authentic Jewish tribes or clans are commonly said to be 
three, Qurayzah, an-Nadir, and Qaynuqa*. This is a simplification, 
however. As-Samhudi has a list of about a dozen clans in addition 
to those already mentioned as being clearly of Arab extraction. 2 
The most important was B. Hadl, closely associated with Qurayzah, 
unless the clan of Tha'labah to which Fityawn belonged is to be 
distinguished from the Ghassanid Tha'labah. Of the three main 
tribes Qaynuqa' possessed no agricultural land but had a compact 
settlement where they conducted a market and practised crafts 
such as that of the goldsmith. Qurayzah and an-Nadir, on the 
other hand, had som of the richest lands in the oasis, situated in 
the higher part towards the south and mostly given over to growing 
palms. Here, as in several other fertile spots in western Arabia 
such as Khaybar, the Jews appear to have been pioneers in agri- 
cultural development. 

The Aws and the Khazraj were allowed to settle, presumably 
on lands that had not yet been brought under cultivation, and were 
under the protection of some of the Jewish tribes. One of the marks 
of their subordinate position was the ius primae noctis exercised by 
Fityawn of B. Tha'labah. Malik b. al-'Ajlan (of B. 'Awf of the 
Khazraj) is said to have been instigated to revolt against Fityawn 
by his sister who wanted to avoid having to spend her first night 
as a bride with Fityawn. Either because he had outside help, or 
because the Jews were temporarily weak owing to outside inter- 
ference, Malik was able to make himself independent. It is difficult, 
however, to estimate accurately the extent of his success. It is 
commonly suggested that the Aws and the Khazraj became rulers of 
Yathrib with all the Jews in subjection to them; but the sources 
do not support such a view. 3 All that we can be certain about is 
that some of the Khazraj became independent; it is probable, 
however, that nearly all the Khazraj and many of the Aws became 
independent, and doubtless acquired strongholds. Indeed, as time 
went on, they seem to have become stronger than the Jews, since 

1 As-Samhudi, 114-16; Wellhausen, op. cit. 12; Aghdm, xv. 162. 16; xix. 
95. 13 f. 2 Cf. Aghdnl, xix. 95. 9 ff. 

3 The general impression is that the Jews were independent. Some of the 
Khazraj are said to be mawdli 'l-yahud in Tab., Tafstr, iv. 22, but this is 

5783 O 


they were able to indulge in the luxury of fighting among them- 
selves. On the other hand, those clans of the Aws which at first 
refused to become Muslim presumably did so because they were in 
close relations with Jewish neighbours. Any estimate of strength, 
of course, must allow for the fact that there were probably serious 
cleavages among both the Arabs and the Jews ; it was improbable 
that there would be a league of all the Jews or all the Arabs. 

Various changes, however, seem to have been taking place in 
the Jewish community which indicate that they were becoming 
relatively weaker. Several Arab clans of the early stratum instead 
of being subordinate to the Jews (doubtless as confederates) be- 
came confederates of Arab clans. Thus B. Unayf became attached 
to B. Jahjaba' (of 'Amr b. 'Awf), and B. Ghusaynah to B. Qawa- 
qilah, 1 while B. Za'ura' had become recognised as members of the 
clan of 'Abd al-Ash'hal b. Jusham, or at least of Jusham, and, as 
confederates of Za'ura', the group known as 'the people of Ratij' had 
become attached to 'Abd al-Ash'hal. 2 By about the time of the 
Hijrah all the lesser Jewish clans or groups in as-Samhudi's list 
had lost their identity, or at least had ceased to be of political 
importance. They are not mentioned in the primary sources for 
the career of Muhammad. When the Constitution of Medina deals 
with them they are simply 'the Jews of an-Najjar', 'the Jews of 
al-Harith', and so on. 3 The nearest to being an exception is Hadl ; 
it had become very closely connected with Qurayzah, but we find 
three members of it becoming Muslims and escaping the fate of 
Qurayzah. 4 From these facts it seems likely that the clan system 
had largely broken down, and that the groups which became 
attached to the various clans of the Ansar were not small clans or 
sub-clans but groups containing people of varying origin. 

The four Jewish clans which feature as clans in the life of 
Muhammad are Qurayzah, an-Nadir, Qaynuqa', and Tha'labah. 
The last must be counted as a Jewish clan, since it seems to appear 
as such in the Constitution of Medina, but it is said to be of Arab 
origin. 5 Qaynuqa' were confederates of 'Abdallah b. Ubayy, and, 
as they supplied him with 700 men (of whom 300 had armour) in 
earlier battles, the parties to the alliance may have regarded one 
another as equals. 6 The other two tribes, an-Nadir and Qurayzah, 

1 Cf. IS, iii/2. 41, 98, &c. 

2 Cf. also the list of Muhammad's opponents in IH, 350 f., and see below, p. 227. 

3 See above, p. 160. 4 IH, 135, cf. 387. 

5 Cf. Wellhausen, Medina, 12, &c. 6 IH, 546; WW, 92. 


were not attacked by Muhammad until after his success against 
Qaynuqa', either because they were stronger, or, more probably, 
because he was more dependent on the support of their Arab 
confederates. The fact that before the battle of Bu'ath they had 
given hostages to the Khazraj suggests that they felt themselves 
at least temporarily weaker than 'Amr b. an-Nu f man al-Bayadi 
and his allies. The affair of the hostages, however, together with the 
rupture between 'Amr b. an-Nu'man and 'Abdallah b. Ubayy, is 
obscure, and does not provide a foundation for a strong argument. 
The incident seems to show the existence among certain of the 
Arabs of a desire to expel the Jews and seize their lands. The two 
clans decided to deal with this threat by exchanging their existing 
alliance for one with the Aws (especially 'Abd al-Ash'hal), even at 
the cost of the lives*of some of their hostages ; thereby they made 
possible the victory of the Aws at Bu'ath. In all this these two clans 
seem to be acting as sovereign bodies, making alliances with Arab 
clans as equals, not politically subordinate to any of them, but 
perhaps tending to become relatively weaker. 

Thus there was little unity among the Jews of Yathrib. In their 
political relationships they behaved in much the same way as Arab 
clans and smaller groups. All had some form of alliance with Arab 
clans, but at least in the case of the stronger Jewish clans this did 
not involve any subordination. By themselves the Jews did not 
constitute a threat to the Arabs, but as supporters of 'Abdallah 
b. Ubayy they might have had considerable influence, and he seems 
to have tried to gain their support. 


There is no mention of any direct negotiations between Muham- 
mad and the Jews before the Hijrah. He must, however, have been 
aware of their importance in Mcdinan politics, and have considered, 
at least provisionally, the attitude he should adopt towards them. 
He believed that the revelation which was coming to him was 
identical with that which had previously been given to Jews and 
Christians. 1 It was natural for him to suppose that this would be 
as obvious to the Jews as it was to him, and that they would 
therefore accept him as a prophet. Presumably some of Muham- 
mad's agents made an approach to the Jews before he left Mecca, 
and their answer may not have been wholly unsatisfactory; they 
1 Cf. Q. 10. 38 C-E+ ; 46. 10/9 ff- DE, &c. 


may have been ready to enter a political agreement, but not to 
accept Muhammad's religious claims. Whatever may have hap- 
pened before the Hijrah, he hoped in his first months at Medina 
to win them over by personal contact. 

The precise status of the Jews in Muhammad's community is 
not clear. They certainly had an indirect relationship to it as con- 
federates of Arab clans which belonged to it. Whether there was 
something more than this it is difficult to say. There is some 
mention of a treaty or covenant in the traditional sources. In one 
passage al-Waqidi says that when Muhammad came to Medina 
all the Jews made an agreement with him, of which one condition 
was 'that they were not to support an enemy against him'; else- 
where he says that the agreement was to the effect that 'they were 
to be neither for him nor against him', and a document was signed 
by Ka'b b. Asad on behalf of Qurayzah, and retained by him until 
the siege of Medina, when it was torn up. 1 In Ibn Ishaq's parallel 
to the latter passage a treaty with Qurayzah is mentioned, but not 
with the other Jews, and nothing that is said implies an actual 
document. 2 A little later the clan says it has no treaty with 
Muhammad, but this might mean either that it had broken its 
treaty or that there never was any. 3 This gives the impression that 
the story has grown in the telling. Ibn Ishaq does not name his 
source; al-Waqidi has two, a grandson of the poet Ka'b b. Malik 
of the clan of Salimah, and Muhammad b. Ka'b (d. 735-8/1 17-20), 
the Muslim son of a member of Qurayzah who escaped death on 
the surrender of the clan since he was only a child; as Salimah 
was hostile to the Jews, and as converts are often bitter against the 
group they have abandoned, both have reasons for making the case 
against Qurayzah as black as possible. Despite some such heighten- 
ing of the melodrama, however, there may be a basis of truth in the 
reports, especially since the terms of the alleged agreement are 
modest and do not imply any close alliance, indeed little more than 
was involved in their being confederates of the Ansar. Such under- 
standing as there was between Muhammad and the Jews may have 
been embodied in a formal document, but it is more likely that the 
Jews were merely mentioned in his agreement with the Ansar (as 
in 1 6 of the extant form of the Constitution) ; the statement in al- 
Waqidi's first account above that Muhammad 'joined each clan 
to its confederates' (sc. of the Ansar) would bear this out. This 

1 WK, 177 (= WW, 92); WW, 196. 2 IH, 674- 3 Ibid. 675. 


would also be sufficient to explain the remark of Abu Bakr in his 
quarrel with Finhas (of Qaynuqa'), 'If it were not for the treaty 
(*ahd) between us and you, I would have cut off your head/ 1 It is 
in accordance with this, too, that the Jews who came to fight at 
Uhud (with the exception of the convert Mukhayriq) are specifi- 
cally said to have done so as confederates of 'Abdallah b. Ubayy. 2 
By way of exception a small number of Jews accepted Muham- 
mad as prophet and became Muslims. The chief of these was appar- 
ently 'Abdallah (originally al-Husayn) b. Sallam (of Qaynuqa'), 
and he was in consequence much maligned by the other Jews. 3 
There was also a group of eight, mainly from Qaynuqa', but they 
seem to have been friends of Ibn Ubayy and became Munafiqun. 4 
Others mentioned by name were converted at a later period, e.g. 
on the day of Uhud,' 5 or at the time of the attacks on an-Nadir 6 and 
Qurayzah. 7 The numbers were sufficient to warrant references to 
them in the Qur'an (at least on the most probable interpretation 
of the passages) ; thus 

among the People of the Book are some who believe in God and in what 
has been sent down to you and in what has been sent down to them, 
humbling themselves to God. . . . 8 

Another passage seems to speak of the Jews who had accepted 
Muhammad as if they were not completely merged with the 
Muslims but formed a separate community (ummah) : 

They arc not all alike ; there is a community of the People of the Book, 
which is steadfast reciting the signs of God at the drawing on of night, 
prostrating themselves, Believing in God and the Last Day, . . . and 
vying in good deeds. 9 

In any case, however, the great majority of the Jews not merely 
did not accept Muhammad, but became increasingly hostile. The 
numerous appeals to the Jews in the Qur'an almost all imply that 
they might be expected to reject the appeals. Very soon after the 

1 IH, 388. 2 WW, 106 (and 124). 

3 IH, 353 f., 387 (quoting Q. 3. 113/109 as referring to converts); cf. Usd, 
iii. 176 f. 4 IH, 361 f. 

5 Mukhayriq (Tha'labah): IH, 354; WW, 124. 

6 Benyamm b. 'Umayr, Abu Sa'd b. Wahb: WW, 164, cf. 98 f. 

7 Asad b. 'Ubayd, &c.: III, 387, 687; (?) Rifa'ah b. Simwal, Usd, ii. 181 (cf. 
IS, viii. 335 f., Ibn Hanbal, Musnad, Cairo, (i895)/i3i3, vi. 37. 26, 193. 5). 
Cf. also WW, 349; IS, i/i. 123. 5-15. 

8 Q, 3- 199/198; cf. 28. 52-55- 


Hijrah it must have become clear that few Jews were likely to accept 
the Gentile prophet. 1 


Early in his career Muhammad must have become aware of the 
similarity between the message that was being revealed to him and 
the teachings of Judaism and Christianity; indeed, according to 
tradition, shortly after the first revelations Waraqah told him that 
what had come to him was identical with the ndmus, presumably 
the Jewish scriptures. Especially after it seemed likely that he 
would go to Medina Muhammad appears to have tried to model 
Islam on the older religion. In the year before the Hijrah, when 
Mus' ab b. 'Umayr was acting as Muhammad's emissary in Medina, 
he asked permission to hold a meeting of th believers, and was 
told he might do so provided he observed the day on which the 
Jews prepared for the Sabbath (that is, Friday, the paraskeue or 
preparation). 2 Thus the Friday worship, which became a distinc- 
tive feature of Islam, was somehow connected with Judaism. 
Muhammad himself does not seem to have observed it until his 
first Friday in Medina. 3 

Another point in which Muhammad may have followed the 
Jewish practice while still in Mecca was in facing towards Jeru- 
salem while worshipping, or, to use the technical term, in taking 
Jerusalem as his qiblah. It is certain that in the early Medinan 
period Jerusalem was the qiblah of the Muslims, but it is doubtful 
whether in Mecca they had this qiblah or another one or none at 
all. The view that even before the Hijrah Jerusalem was the qiblah 
of at least the Medinan Muslims is supported by a story about 
al-Bara' b. Ma'rur, the leading Muslim of B. Salimah. During the 
expedition to Mecca shortly before the Hijrah he refused to turn 
his back on the Ka'bah, though his companions expostulated. 
In Mecca Muhammad was consulted, and told him to return to 
his previous qiblah, Syria (that is, Jerusalem). 4 Muhammad him- 
self may have had no qiblah at this time but may have been 

1 For umml as 'Gentile' see Horovitz, Koranische Untersuchungen, 51-53; 
R. Paret, art. 'Ummi' in EI(S), with references; also H. L. Fischer, Kleinere 
Schriften, Leipzig, 1888, ii. 115-17. 

2 IS, iii/i. 83. 23 fT., reading yatajahhazu cf. C. H. Becker in Der Islam, 
*" 379; Wensinck, op. cit. 111-14; Buhl, Mohammed, 214. 

3 Tab. 1256. 20; cf. Caetani, i. 375 f. (writing without access to IS, iii/i). 

4 IH, 294 f.; Tab. 1218 f. 


keen to assimilate his religion to that of the Jews in this respect. 
If in his Meccan period Muhammad faced towards Jerusalem, 
this would not necessarily indicate Jewish influence or a desire 
to be like the Jews, since the practice was apparently common 
among Christians. l By 624/2, however, he was aware that there were 
differences between Jews and Christians. 2 On the whole it seems 
most likely that the Jerusalem qiblah was adopted by Muhammad 
from the Medinan Muslims, especially since the main point in the 
story about al-Bara' is confirmed by the Qur'an (2. I50/I45). 3 

There is less uncertainty about the institution of a fast on the 
Jewish Day of Atonement, the Fast of 'Ashura. 4 Muhammad 
certainly commanded the Muslims to observe this fast when the 
loth of the Jewish month of Tishri came round, though it is not 
certain in which of the Muslim months this fell. Perhaps some of 
the Medinan Muslims had already been in the habit of observing 
it, for, when the fast of Ramadan was instituted, that of the f Ashura 
was not forbidden, though it ceased to be obligatory. Similarly, 
in accordance with Jewish practice, midday worship (saldt) was 
instituted. In Mecca there had apparently been only morning and 
evening worship, apart from nocturnal vigils; 5 but at Medina the 
Qur'an commands, * Remember the Worship, the middle Worship 
included' (or, 'especially the middle Worship'). 6 Further, it has 
been suggested that in building the mosque at Medina Muhammad 
had in mind the Jewish synagogue ; but what became the mosque 
was primarily his own house and courtyard, and there are strong 
grounds for doubting any imitation of an ecclesiastical building. 7 
Apart from this, however, there is evident in Muhammad, shortly 
before and shortly after the Hijrah, a tendency to make his religion 
similar to that of the Jews and to encourage his Medinan followers 
to continue Jewish practices which they had adopted. 

The same aim of reconciling the Jews probably underlies the 
verse which permits Muslims to eat the food of the People of the 

1 Tor Andrae, Ursprung des Islarns, 4; Buhl, 218. 

2 Q. 2. 145/140; cf. Bell, Origin of Islam, 144. 

3 Cf. discussions of the whole question by Wensinck, op. cit. 108-10, and 
Buhl, 216-18; also D. G. Margoliouth, in Journal of the Royal Asiatic Society, 

1925, 437- 

4 Q. 2. 183/179; Tab. 1281 ; cf. Wensinck, 122-5; Buhl, 214; Caetani, Ann. i. 
431 f., 470 f. 

5 Cf. Q. ii. 114/116; Buhl, 215; Wensinck, 106-8. 6 2. 238/239. 

7 Buhl, 204 f.; Caetani, i. 432 ff. ; Wensinck, 116; C. H. Becker, Islamstudien, 
Leipzig, 1924, i. 450 (= Noldeke-Festschrift, Giessen, 1906, 331). 


Book and to marry women from them (5. 5/7). This presumably 
refers to the Jews. It may be that the Meccan Muslims did not at 
first realize that the Jews had numerous restrictions, and thought 
that only the things mentioned in the Qur'an were forbidden, 
namely blood, pork, and animals that had died naturally or been 
strangled or sacrificed to idols. 1 (It is curious that this list, apart 
from the mention of pork, should be so like that in Acts xv. 18; 
and one wonders whether this represents a common level of obser- 
vance among monotheists in the Arabian peninsula, both Jews of 
Arab descent and Christians.) At any rate, there is no record of 
Muhammad ever expecting his followers to observe all the Jewish 
restrictions ; and after his break with the Jews we find in the Qur'an 
denials that these are part of the revelation to the Jews from God, 
and the suggestion that they are intended as 3 punishment for the 
Jews. 2 Some of the phrases of the Qur'an, however, might imply 
that Muslims had been following Jewish practice. 3 

All such ordering of the new religion to make it conform more 
closely to the older one was probably inspired by two motives, 
the desire for a reconciliation with the Jews and the desire to sub- 
stantiate the reality of Muhammad's prophethood by showing the 
essential identity of his revelation with the preceding one. Latterly 
the second motive may have become dominant, but to begin with 
the other must also have been prominent. Indeed, there are slight 
traces of Muhammad's being ready to make far-reaching concessions 
to Jewish feeling. In the first section of this chapter it was assumed 
(as Western scholars have usually assumed) that Muhammad's 
appeal to the Jews was an appeal to become Muslims or rather 
'believers' on exactly the same footing as his Arab followers. It was 
noticed, however, that he could also speak of certain Jews who had 
apparently responded to his appeal as forming a distinct com- 
munity. When this latter point is linked up with others about to 
be mentioned, there is some justification for thinking that at some 
period during the first year or so at Medina (not necessarily in the 
first months) Muhammad contemplated a religious and political 
arrangement which would give a measure of unity but would not 
demand from the Jews any renunciation of their faith or acceptance 
of Muhammad as prophet with a message for them. Such an 

1 Q- 5- 3/4; cf. 16. 115/116. 

a 3. 78/72 F; 4. 160/158 F; 6. 146/147 E-f ; 10. 59/60 c-E-f ; 16. 118/119 E. 

3 16. 116/117 E, &c. (apparently addressed to believers). 


arrangement would be in accordance with the general idea that 
each prophet was sent to a particular community, and that the 
community to which he was sent was the Arabs. There seems to 
be such an appeal for reconciliation on the basis of monotheism 
and nothing else in a verse (possibly revealed in A.H. 2) : 

Say: *O People of the Book, come to a word (which is) fair between 
us and you, (to wit) that we serve no one but God, that we associate 
nothing with Him, and that none of us take others as Lords beside 
God. 1 

Moreover the verse permitting Muslims to eat the food of the 
People of the Book (5. 5/7) takes on a different colour if it is regu- 
lating the relations of two religious groups within a single political 
community. This would explain how it comes about that it 
appears to legislate for the Jews by making Muslim food permis- 
sible for them an act which the Jews would regard as one of 
presumption. There is no mention in extant records of Jews marry- 
ing Muslim women, either because there were not sufficient Muslim 
women, or because there was a mention but it dropped out when 
the practice ceased. If there was thus a Jewish ummah as well as 
a Muslim ummah within the one political entity, it is conceivable, 
though not probable, that in the phrase of the Constitution about 
the Emigrants and the Ansar forming an ummah * distinct from 
the (other) people' (dun an-nas), the word nds or 'people* refers not 
to people in general but to the People par excellence. 21 

Despite the concessions Muhammad was prepared to make and 
his attempt to render his religion similar to that of the Jews, 
the latter did not become any friendlier towards him as time 
went on. On the contrary, they became hostile, and broadcast 
adverse criticism of Muhammad's claims to be a prophet. Their 
reasons for this may have been partly religious the obvious 
contradiction between what Muhammad claimed or asserted and 
some of their fundamental dogmatic attitudes. There is no way, 
however, of measuring the strength of this motive, for the matter 
had a political aspect, and this also, one may suspect, was of im- 
portance. If Muhammad succeeded with his plan, the Jews would 
have no chance of supreme power; they may have realized al- 
ready that the Emigrants would generally have more influence on 

1 3- 64/57; for the last clause cf. 9. 31. 

2 Constitution, i; cf. 15. 


Muhammad than the Ansar. On the other hand, until the battle of 
Badr Muhammad's prospects of success were poor, and they may 
have thought that they would be better off if there was a return 
to the status quo\ for some of them hopes may have been set 
on a league with Ibn Ubayy. They were, of course, far from being 
united, and their motives doubtless varied from clan to clan. All, 
however, with the few exceptions noted above, rejected Muham- 
mad's appeals. 

Muhammad remained outwardly patient for some time. Then 
there was a sudden change in his attitude at least if we may 
believe a story not found in the earliest sources. One day while 
engaged in the Worship at the prayer-place in the quarter of 
B. Salimah, he received a revelation bidding him turn from facing 
Syria and face the Ka'bah instead. He did so,, followed by the other 
participants, and the spot became the site of the Mosque of the 
Two Qiblahs. 1 A less colourful but more likely version is that he 
received the revelation of 2. 144/139 by night, and communicated 
it to the believers the following day. 2 The date is usually given as 
about the i5th of Sha'ban, A.H. 2 ( = n February 624). 3 On the 
other hand, the verses referring to a change of qiblah (2. 142/136 
152/147) show different strands and must have been revealed at 
different times. Richard Bell in his Translation therefore suggests 
that there may have been an interval between dropping the Jeru- 
salem qiblah and adopting the Meccan. Certainly there seems to 
have been a period of hesitation. There is a report that the Jews 
taunted the Muslims that they did not know where to turn in 
worship until they (the Jews) told them, and that this made 
Muhammad desirous of the change. 4 The Muslims, too, may have 
been divided among themselves. It is significant (though perhaps 
only for the criticism of sources) that the change of qiblah is said 
to have taken place among B. Salimah, for this was the tribe of 
al-Bara' b. Ma'rur, who before the Hijrah was an advocate of the 
Ka'bah as qiblah. If the traditional date is correct, the change must 
have been made about the time of the raid of Nakhlah by which 
a challenge was issued to Quraysh; it was also just before Badr. 

About the same time traditionally in the same month of 
Sha'ban (viii), but more probably in the following month of 

1 Ad-Diyarbakri, al-Khamis, i. 414. 17-20. 

2 Al-Bukhari, $aldt (8), 32; Tafsir (65), on 2. 144/139. 

3 IH, 427, contrast 381; Tab. 1279 f. 4 Ibid. 1281. 


Ramadan (ix) after Badr, about the igth ( = 15 March) Muham- 
mad instituted the fast of Ramadan and declared that of the 
' Ashura no longer obligatory. 1 Various suggestions have been given 
for the source of this new practice, from the Christian Lent to 
customs of Manichaeans and pre-Islamic Arabs. 2 Most light on its 
significance for Muhammad is thrown by Bell's view that the 
victory at Badr was the Furqdn, that is, the coming of the promised 
calamity upon the unbelievers and the deliverance of the believers, 
analogous to the deliverance of Moses at the Red Sea, and that 
in commemoration of this Furqan the month's fast was instituted. 3 
There is some confirmation for this view in the account of at- 
Tabari: 4 

In this year, according to report, the fast of the month of Ramadan 
was instituted ; the institution is said to have been in Sha'ban ; when the 
Prophet (God bless and preserve him) came to Medina, he observed 
the Jews fasting on the day of 'Ashura, and questioned them; they 
informed him that it is the day on which God caused the drowning of 
the host of Pharaoh and delivered Moses and those of them who were 
with him; Muhammad remarked, We have more right than they, and 
both fasted himself and bade the people fast on that day ; when the fast 
of the month of Ramadan was instituted, he neither commanded them 
to fast on the day of 'Ashura nor forbade them. 

No isnad is given for this, but it may nevertheless contain in 
a slightly distorted form the memory of how Muhammad had 
originally connected the fast with the victory on the analogy of 
a supposed connexion between the Jewish fast and the deliverance 
of Moses from the Egyptians. The chief difficulty about Bell's 
view is the date. At-Tabari's mention of Sha'ban, however, is 
hesitating, and is presumably an inference from the fact that the 
fast would have to be proclaimed before it was due to begin. 
Muhammad is said to have fasted a day or two on the way to Badr, 
though it was permitted to those on a journey not to fast. 5 As this 
was the first occasion of the fast, however, one would expect the 
matter to be treated more explicitly. It is difficult to resist the 
conclusion that the fast of Ramadan was not fully observed before 
A.H. 3. 

1 Ibid. 

2 Caetani, i. 470 f.; Wensinck, 137; Buhl, 227, with further references. 

3 Origin of Islam, 124 f. 

4 Tab. 1281 (not referred to explicitly by Bell), omitting the words 'to Moses* 
after 'right', following one manuscript. 5 WW, 46. 


These marks of 'the break with the Jews' are in fact indications 
of a completely new orientation both politically and religiously. 1 
The Medinan state now began a series of attacks on the Jews in 
the physical sphere, and at the same time the Qur'an carried on 
polemics against their religion in the intellectual sphere. 

The reasons for the new policy are not far to seek. So long as 
Muhammad claimed to be receiving revelations identical in essence 
with the revelation in the hands of the Jews, they were in a strong 
position, and could either support Muhammad by acknowledging 
the similarity or hinder his cause by drawing attention to differ- 
ences. It was mostly the latter that they chose to do, and conse- 
quently they threatened to undermine the intellectual foundations 
of his political and religious position. Muhammad was always very 
sensitive to such ideological attacks, and, for example, dealt severely 
with poets who opposed him. His stern attitude towards the Jews 
when they rejected his appeals was not simply pique at this rejec- 
tion, but the reaction of a man in danger to those whose ill will is 
causing this danger. 


In the polemics of the Qur'an against the Jews a prominent 
place is taken by the conception of the religion of Abraham. This 
is an idea which is not found in the Meccan revelations and is 
presumably not based on pre-Islamic Arab legends. - During the 
Meccan period more prominence was given to Moses than to 
Abraham among the prophets as a forerunner of Muhammad. 
Abraham is simply one of many prophets, and the people to whom 
he is sent are not specified; indeed, it seems to be implied that he 
was not sent to the Arabs, since Muhammad is said to be sent to 
a people who had never had a warner. 2 Likewise there is no mention 
of any connexion of Abraham and Ishmael with the Ka'bah; 
Ishmael is named in lists of prophets, but no details are given about 
him. 3 The presumption is that at first the Muslims did not know 

1 Cf. Buhl, 228; and for further points of detail, D. G. Margoliouth, Moham- 
med and the Rise of Islam, London, 1905, 250. 

2 32. 3/2 E-f ; 34. 44/43 D-E; 36. 6/5 c. The argument is a repetition of that of 
C. Snouck Hurgronje, Verspreide Geschriften, Bonn, &c., 1923-7, i. 22-29, 334-8 ; 
cf. Buhl, 229-31 ; Bell, Origin, 129-31 ; contrast E. Beck in Museon, Ixv (1952), 
73-94. (The first passage from Snouck Hurgronje has been translated into 
French and annotated by G.-H. Bousquet, Revue Africaine, xcv (1951), 273-88.) 

3 6. 86 ? C-B+ ; 21. 85 ? D; 38. 48 CD. 


about the connexion of Ishmael with Abraham and (according to 
the Old Testament) with the Arabs. At Medina, however, in closer 
contact with the Jews they gained knowledge of such matters. 
When it came to a break with the Jews, Abraham had two great 
advantages : he was in a physical sense the father of the Arabs as 
well as of the Jews ; and he lived before the Torah had been revealed 
to Moses and the Gospel to Jesus (as the Jews had to admit), and 
was therefore neither a Jew nor a Christian. 

The Qur'an therefore instructs Muhammad and the believers 
to regard themselves as neither Jews nor Christians, but a com- 
munity distinct from both, followers of the 'creed of Abraham* 
(millat Ibrahim); and Abraham is described as a hanif, a muslim 
(that is, one surrendered to God), not one of the idolaters. 1 The 
religion of Abraham is simply the pure religion of God, since all 
the prophets have received in essentials the same revelation. 
Judaism and Christianity, however, now come to be looked on as 
imperfect manifestations of this religion of God, and therefore 
a distinctive name has to be found for it. First hanif and later also 
muslim are used in the Qur'an for the adherent of the true religion; 
hanif had apparently been used previously by Jews and Christians 
either for 'pagan* or for 'a follower of the Hellenized Syro- Arabian 
religion', and is thus given a completely new turn of meaning by 
the Qur'an; 2 muslim is presumably a new coinage. Moreover, 
Abraham is now said to have founded the Meccan sanctuary with 
the help of Ishmael, and to have prayed for a prophet for the 
Meccans from among his descendants, 3 while in addressing the 
believers the phrase 'your father Abraham' is used. 4 In these ways 
the practice of facing Mecca during the Worship comes to be 
supported by a vast ideological structure. 

The corollary of the conception of the religion of Abraham is 
that the religion of the Jews is not the pure religion of Abraham. 
This idea is implicit in a number of passages, and becomes explicit 
in certain specific points. One of these is that the Jews have broken 
their covenant with God, made at Sinai, by worshipping a calf 
instead. 5 Another is that they disbelieve in part of the Book which 
has been given to them, and act wickedly in disobedience to the 

1 2. 130/124-141/135; 3. 65/58-71/64; 5. 44/48 f.; 6. 159/160-165; 14. 35/38; 
16. 120/121-123/124. 

2 Cf. Ml Mecca, 162 ff. and references. 

3 2. 125/119-129/123. 4 22. 78/77. 

5 2. 27/25, 40/38, 83/77; 4- 154/153 f-; 5- 12/15 f.; &c. 


commands of God. 1 The taking of usury is an instance of such 
disobedience. 2 In all this they show their worldliness. 3 Moreover 
all that they allege to belong to the Book revealed to them is not 
in fact part of that Book ; such statements apparently refer to the 
Jewish oral law, and they would serve to explain the absence of 
corresponding regulations in the revelation to Muhammad. 4 Not 
dissimilar to this is the charge of 'altering words from their proper 
meanings' (yuharrifuna 'l-kalim 'an mawddii-hi)? in the Qur'an 
this need mean nothing more than deliberately interpreting pas- 
sages to suit oneself, and neglecting the plain and straightforward 
meaning; but later Muslim apologetic took this to mean that the 
Jewish and Christian scriptures were textually corrupt. 

The repeated assertions that the Jews conceal part of the truth 
revealed to them point to the corruption of individual Jews and not of 
their religion, but there may be a close connexion with the charge of 
* altering* or * making oblique' (commonly referred to as tahrlf, this 
being the noun corresponding to yuharrifuna). Thus in 2. 76/71, im- 
mediately after some Jews have been accused of tahrlf, there is a 
description of how some who feign to be believers (that is, to accept 
Muhammad) say to one another in privacy that it is foolish to tell the 
Muslims what has been revealed to them (the Jews) since the Mus- 
lims will be able to produce it as proof against them in the presence 
of God. In certain other passages the reference might be to 
the concealing of such facts as that Abraham was the father of the 
Arabs and was not a Jew, but there can be little doubt that what 
is concealed in the verse under discussion is the fact that Muham- 
mad and his prophetic mission are foretold and described in the 
Torah (as stated elsewhere in the Qur'an). 6 In this verse, then, the 
Jews are represented as knowing that Muhammad fulfils the scrip- 
tural description of the prophet who is to come (as did the convert 
'Abdallah b. Sallam, according to Ibn Ishaq); 7 but these Jews, 
though they profess to believe, are not prepared to act accordingly 
and become whole-hearted followers of Muhammad, and they will 
therefore be severely treated on the Last Day though of course, 

1 2. 85/79; 5- 78/82 f.; 62. 5. * 4- 161/159- 

3 2. 86/80, &c. 
* 3. 78/72, 93/87- 

5 4. 46/48; 5. 13/16; cf. 5. 41/45; 2. 75/7o* and Lane, s.v. harafa. 

6 7- I57/IS6; cf. 4. 37/41 (with Tab., Tafsir, v. 51); IH, 388. For the main 
point cf. comment of 'Abd al-Qadir in Wherry (Sale) on 2. 76/71 (i. 317). 

7 IH, 353; contrast Kinanah b. Suriya' in WW, 161. 


as the next verse states, God knows what they are keeping secret 
and does not require to be informed about it by the Muslims. In 
other verses of the Qur'an the Jews are simply said to conceal 
truth, 1 but no doubt the reference is mostly to the description of 
Muhammad as prophet. (Ibn Ishaq's story of how some Jews tried 
to conceal the verse in the Torah which made stoning the punish- 
ment for adultery belongs rather to later controversies and cannot 
be used uncritically to determine the meaning of the Qur'an.) 2 

So long as the Muslims knew little about the Jewish scriptures, 
it was possible for the Jews to get the better of most arguments. 
But, with growing knowledge, the Muslims were able to use the 
scriptures against the Jews. The point that Abraham was not a 
Jew has already been mentioned. In particular, the Torah provided 
excellent material fo? countering the Jewish rejection of Muham- 
mad. Some of his followers had probably been perturbed when 
they saw how the Jews, whom they respected in religious matters 
as People of the Book, did not acknowledge him. The force of this 
consideration, however, was greatly weakened by showing that this 
was no new feature of Jewish history, but that their sacred record 
was full of instances of their rejection of those who came to them 
from God. 

If then they count thee false, messengers have already before thy 
time been counted false who came with the Evidences and the Psalms 
and the illuminating Book. 3 

It was further suggested that the Jews' rejection of Muhammad 
was not based on their scriptures but was due to base motives such 
as envy or jealousy. 4 

Much of the Jewish strength presumably lay in their absolute 
conviction that they were God's chosen people. Some of the more 
presumptuous forms of this conviction are described in the Qur'an. 
They hold that they are 'justified', and that they alone will be in 
Paradise; if they go to Hell at all, they will only be a limited time 
there. 5 To such claims, which were tantamount to a dismissal of 
Islam as completely false, the Qur'an had various forms of reply. 
It could deny directly, as when it insisted that the judgement 

1 2. 42/39, 146/141, 159/154, 174/169; 3. 71/64; 5. 15/18; 6. 91. 

2 IH, 394 f- 

3 3- 184/181; cf. 2. 61/58, 87/81, 91/85 f.; 3. 181/177; 4. 155/154; 5- 70/74 f- 

4 2. 105/99, 109/103; Cf. 2. 90/84; 4. 54/57. 

5 4- 49/52; 2. 94/88, 80/74; 3- 24/23. 


passed on the Last Day depended on a man's righteousness and 
obedience to God's commands. 1 It could deny by implication, as 
when it asked why the Jews (and Christians) were punished, if, as 
they held, they were the sons of God. 2 In accordance with old Arab 
custom they could be challenged to take an oath to the effect that 
they were the friends of God and that they alone would be in 
Paradise. 3 But perhaps the trump card was that Jews and Christians 
denied one another's exclusive claims. The two claims were similar 
and therefore could not both be true; but there was little to choose 
between them, and it was thus not unreasonable from the Muslim 
standpoint to suppose that both went beyond what their revealed 
scriptures justified. This point had the more force in that Jews and 
Christians were apparently regarded by the Arabs as being two 
branches of the Children of Israel. With regafd to the matters about 
which they differed the decision had been postponed by God until 
the Last Day. 4 As against the exclusive claims of the two older 
religions there was a show of broadmindedness and tolerance in 
the Muslim claim to acknowledge all earlier prophets, 5 and the 
idea of a covenant between God and the prophets, in which they 
promised to believe in and help any subsequent prophet with a 
message confirming the existing revelation, might be regarded as 
setting forth an important truth in mythic form. 6 

Such are the main points of the Qur'anic attack upon the Jews. 
There were also some minor matters, but sufficient has been said to 
show that the Muslim attitude towards the Jews was well developed. 
This degree of elaboration is an index of the great importance of 
the Jewish question for the Muslims. In the thoughts of their 
leaders it must have bulked at least as largely as the struggle with 
Mecca. This has to be kept in mind in considering the actual 
hostilities between Muhammad and the Jews. 


During the months and years that followed the change of the 
qiblah there were a number of hostile encounters between the 
Muslims and the Jews. While it is convenient to group these to- 
gether, it should not be assumed without examination that all these 
events spring from a deliberate policy presumably adopted in 

1 2. 80/74 #; 3- 24/23 f. a 5. 18/21. 3 2. 94/88; 62. 6 ff. 

4 2. 111/105 ff.; cf. 10. 93; 27. 76/78; 42- 14/13 ; 45- 16/15 ff.; 98. 4/3* 

5 Cf. 4. 150/149 ff.; 2. 136/130. 6 3. 81/75. 


623/2 before Badr of subduing or getting rid of the Jews. Whether 
this is so is a question that must be discussed, but it may be post- 
poned until the events themselves have been briefly described. 

The first event of note was the siege and expulsion of the clan of 
Qaynuqa'. 1 On his return from Badr Muhammad is said to have 
renewed his appeals to the Jews, pointing to the Meccan losses as 
an example of the fate of those who did not respond to God's 
message. The Jews, however, were no readier than before to become 
followers of Muhammad. A few days later an incident occurred. 
Some Jews played a trick on an Arab woman. 2 While she was sitting 
doing business in the market of the Qaynuqa', one of them con- 
trived to fasten her skirt in such a way that when she stood up 
a considerable portion of her person was revealed. A Muslim who 
happened to be prese'nt regarded this act and the ensuing laughter 
as an insult, and killed the Jew, who was at once avenged by his 
fellows. The Jews then retired to their strongholds. Muhammad 
regarded the matter as a casus belli, and collected a force to besiege 
the clan. There were doubtless some negotiations, but no record 
has been preserved. After a siege of fifteen days the Jews sur- 
rendered. They were forced to leave Medina, taking their wives 
and children with them. Three days were granted to them to 
collect money owing to them, but they had to leave behind their 
arms and perhaps some of their other goods, such as their gold- 
smith tools (though one might conjecture that by the latter are 
meant the tools used in making weapons and armour). The usual 
account is that they went to the Jewish colony at Wadi '1-Qura, 
and after a month proceeded to 'Adhra'at in Syria. 

It is important to notice the part played by various Arabs in 
these happenings. Ibn Ubayy receives most prominence, since he 
spoke to Muhammad on behalf of Qaynuqa' ; and they are said to 
have become reconciled to exile only after they saw that Ibn Ubayy, 
their confederate, on whose support they were counting, had little 
influence in Medinan affairs ; when he tried to force his way into 
Muhammad's presence the man on guard pushed him so violently 
against the wall that his face bled, and Ibn Ubayy was apparently 
incapable of exacting revenge or compensation. Others, however, 
are also mentioned as prominent in the operations on the Muslim 
side, namely, 'Ubadah b. as-Samit, al-Mundhir b. Qudamah, 

' IH, 545-7J WK, 177-81; WW, 92-94; Tab. 1360-2. 
2 For a similar trick, cf. Caussin de Perceval, i. 297 f. 

6783 P 


Muhammad b. Maslamah, and Sa'd b. Mu'adh. The first two 
belonged to branches of the old clan of Salim, 1 al-Qawaqilah and 
Ghanm b. Silm respectively. IbnUbayy was of Ba'1-Hubla, another 
branch of Salim. As 'Ubadah also had had a confederacy with 
Qaynuqa' but had publicly denounced it when tension began to 
grow, it may be inferred that Qaynuqa' had originally been con- 
federates of the whole of Salim. The other two Arabs were of 
the clan of 'Abd al-Ash'hal which had formed an alliance with 
an-Nadir and Qurayzah just before the battle of Bu'ath and may 
therefore have had some relationship to Qaynuqa' also. Alterna- 
tively, if there was no such relationship, Sa'd b. Mu'adh's con- 
tribution may have been to keep the other Jewish clans from 
interfering. Qaynuqa' are said to have had 700 fighting men, of 
whom 400 wore armour, and Muhammad coiild not have been suc- 
cessful against them without the whole-hearted support of many of 
their confederates among the Arabs. His high prestige after Badr 
no doubt made it easier for him to gain such support. 

Four months or so later (early September, 624 the middle of 
iii/3), there occurred the assassination of Ka'b b. al-Ashraf. 2 Ka'b 
was the son of an Arab of the distant tribe of Tayyi', but he behaved 
as if he belonged to his mother's clan of an-Nadir. After Badr he 
went to Mecca and composed anti-Muslim verses which had a wide 
circulation. At Muhammad's instigation the Muslim poet, al- 
Hassan b. Thabit, satirized Ka'b's Meccan hosts and so forced 
him to return to Medina, where he continued his propagandist 
activities. Muhammad apparently let it be known that he would 
gladly be rid of Ka'b and, when five men hatched a plot against 
him, gave them permission to say what they liked about himself. 
Two, Muhammad b. Maslamah of the section of Harithah attached 
to e Abd al-Ash'hal and Abu Na'ilah of the sub-clan Za'ura of the 
same clan, were milk-brothers of Ka'b and one or other of them 
secured his confidence by complaining of the hardships they had to 
suffer under Muhammad's regime, and in particular of the lack of 
food. Ka'b agreed to give them a loan and to accept arms as a pledge. 
To receive the arms he left his house in the middle of the night. 
All five set upon him at a quiet spot and not without some difficulty 
killed him. On their return to within earshot of Muhammad, who 
was watching for their return, they announced their success by 

1 Cf. p. 167 above. 

* IH, 548-53; WK, 115-17, 184-90; WW, 74, 95-99J Tab. 1368-72. 


a shout of Allah akbar^God is very great'. It is noteworthy that the 
five conspirators were all members of ' Abd al-Ash'hal or the closely 
connected Harithah. As an- Nadir were confederates of 'Abd al- 
Ash'hal, no blood-feud would be created. The Jews are said to have 
been greatly perturbed at the assassination, to have complained to 
Muhammad, and to have entered into a treaty with him. 

Almost exactly a year after KaVs death, in iii/4 ( late 
August or early September 625), a second Jewish clan, Banu 'n- 
Nadir, were expelled from Medina. 1 The story is that Muhammad 
went to the settlement of an-Nadir to demand a contribution, 
towards the blood-money due to B. 'Amir b. Sa'sa'ah for the two 
men killed by the survivor of Bi'r Ma'unah. 2 As an-Nadir were in 
alliance with 'Amir, there may have been complications, though 
the sources say nothing of these ; Muhammad may have thought 
that the Jews ought to do more than the average of the inhabitants 
of Medina, and they may have thought they ought to do less. What- 
ever the precise point was, an-Nadir professed themselves ready 
to give a satisfactory answer, but bade Muhammad make himself 
comfortable while they prepared a meal. He and his companions 
seated themselves with their backs to the wall of one of the houses. 
Presently Muhammad slipped quietly away and did not return, 
and his companions also eventually left. When they found him at 
his house, he explained that he had had a Divine warning that 
an-Nadir were planning a treacherous attack on him they could 
easily have rolled a stone onto his head and killed him as he sat by 
the house. He therefore at once dispatched Muhammad b. Mas- 
lamah to an-Nadir with an ultimatum ; they were to leave Medina 
within ten days on pain of death, though they would still be re- 
garded as owners of their palm-trees and receive part of the 
produce. Such an ultimatum seems out of proportion to the offence, 
or rather to the apparently flimsy grounds for supposing that 
treachery was meditated. Yet perhaps the grounds were not so 
flimsy as they appear at first sight to the Westerner of today. Both 
parties knew how some Muslims had treated Ka'b b. al-Ashraf, 
and, in accordance with the ideas of the Arabia of that day, Mu- 
hammad was bound to expect that, if he gave his opponents an 
opportunity, they would kill him. An-Nadir's postponement of 

1 IH, 652-6; WK, 353-62; WW, 160-7; Tab. 1448-53- 

2 For Abu Rafi' see IH, 714-16, 981; WW, 170-2; Tab. 1375-83; Caetani, 
i. 590-2. For Usayr cf. IH, 980 f.; WW, 239 f.; Caetani, i. 702 f. 


a reply created such an opportunity, and was therefore tantamount 
to a hostile act. 

The Jews at first were inclined to submit to the demand, espe- 
cially when they saw that it was carried by a leading member of 
the clan on which they were primarily dependent for support. They 
were divided among themselves, however. Huyayy b. Akhtab, 
apparently chief of the clan, was less inclined to submit than other 
men such as Sallam b. Mishkam. While Huyayy hesitated, Ibn 
Ubayy sent messages to him promising support and speaking of the 
readiness of some of the allied nomads to attack Muhammad. The 
Jews therefore refused to comply with Muhammad's demand, and 
he set about besieging them. The siege lasted about fifteen days. 
An-Nadir lost heart when the Muslims began to destroy their 
palms, for Ibn Ubayy was doing nothing to* help them and they 
realized that, even if they were able to keep their foothold in 
Medina, their livelihood would be gone. They expressed their 
readiness to fulfil the original demand, but Muhammad now im- 
posed less favourable terms on them. They were to leave their 
weapons and to have nothing from the palms. To this perforce they 
agreed, and departed proudly with a train of 600 camels for Khay- 
bar, where they had estates. The swords, cuirasses, and helmets 
all went to Muhammad, doubtless with a view to his next encounter 
with Quraysh. The Ansar agreed that the houses and palm-gardens 
should be allotted to the Emigrants, so that they mighj; be able to 
support themselves and be no longer dependent on the hospitality 
of the Ansar. Among the Muslims mentioned in connexion with 
the affair, Muhammad b. Maslamah and Sa'd b. Ma'adh are 
prominent, but it is significant that Sa'd b. 'Ubadah provided 
a specially fine tent for Muhammad and dates for the whole army. 
This may indicate that he was coming forward as leader of all the 
Khazraj in opposition to Ibn Ubayy. Of the two poor Ansaris who 
shared with the Emigrants in the distribution of the confiscated 
property, one was Abu Dujanah from Sa'd b. 'Ubadah's clan of 

The expulsion of an-Nadir from Medina was not the end of their 
dealings with Muhammad. From Khaybar some of them continued 
to intrigue assiduously against Medina, and they played a consider- 
able part in the formation of the great confederacy to besiege 
Medina in April 627 (xi/5). It is not surprising, therefore, that two 
of their leaders, Abu Rafi' Sallam b. Abi '1-Huqayq and Usayr (or 


Yusayr) b. Razim, were assassinated by Muslims. The dates 
adopted by al-Waqidi are respectively xii/4 ( =May 626) and 
x/6 ( = February-March 628), but there are variants, notably some 
which place the former after the siege of Medina and the attack 
on Qurayzah. 1 This later date seems to be slightly more probable. 2 
Al-Waqidi 3 states that Usayr b. Razim became leader in war of an- 
Nadir after the death of Abu Rafi e . If this position is identical with 
that occupied by Huyayy b. Akhtab, then Abu Rafi* could not have 
assumed it until after the .death of Huyayy along with Qurayzah, 
and could not have been assassinated until after that. The reason 
for the assassination of Abu Rafi* as for that of Usayr was intrigues 
with Ghatafan against the Muslims, and this would fit either date. 

The attack on Abu Rafi' was the work of five men of B. Salimah. 
They are said to have been moved to it by the desire to show that 
the dispatching of Ka'b b. al-Ashraf by the Aws could be rivalled 
by the Khazraj. A prominent part was played by 'Abdallah b. 
* Atiq, who spoke Hebrew and had a Jewish foster-mother in Khay- 
bar (perhaps a woman of an-Nadir); but the leader and person 
chiefly responsible for the actual assassination appears to have been 
'Abdallah b. Unays. The party managed to gain admittance to the 
house of Abu Rafi', and had little difficulty in mortally wounding 
the old man. They hid until the pursuit died down and then re- 
turned safely to Medina. 

f Abdallah b. Unays seems to have been responsible for the killing 
of Usayr also, though the leader of the party of thirty was 'Abdallah 
b. Rawahah (of Ba'1-Harith). They went openly to Khaybar as 
representatives of Muhammad with talk of honours to be bestowed 
on Usayr and an invitation to a parley in Medina. Despite warnings 
from some of his friends, Usayr and thirty companions set off for 
Medina, each mounted behind one of the Muslims. On the way 
'Abdallah b. Unays became suspicious of Usayr, who was behind 
him and seemed once or twice to be feeling for 'Abdallah's sword, 
presumably regretting his decision; from this and from the fact 
that later he used the branch of a tree it is to be inferred that the 
Jews were unarmed. 'Abdallah contrived that his camel lagged 
behind the others and when they were alone killed Usayr. Subse- 
quently the other Jews were also killed with one exception. 

1 For Abd Ratt see IH, 714-16, 981; WW, 170-2; Tab. 1375-83; Caetani, 
i. 590-2. For Usayr cf. IH, 980 f.; WW, 239 f.; Caetani, i. 702 f. 

2 Cf. Buhl, 277, n. 48. 3 WK, 4. 17; WW, 239 n. 


There were still a number of Jewish groups in Medina, but the 
only one of any importance was the clan of Qurayzah. During the 
siege of Medina this clan had probably preserved neutrality so far 
as outward acts were concerned, but they had engaged in negotia- 
tions with Muhammad's enemies, and, could they have trusted 
Quraysh and their bedouin allies, would have turned against 
Muhammad. Immediately upon the withdrawal of his opponents 
Muhammad attacked Qurayzah, 1 to show that the rising Islamic 
state was not prepared to tolerate such 'sitting on the fence*. Quray- 
zah retired to their stronghold, but did not fight back with much 
vigour. Soon they sent and asked to be allowed to surrender on the 
same terms as an- Nadir, but were told they must surrender uncon- 
ditionally.They then requested to be allowed to consult Abu Luba- 
bah, and he went to them. What exactly happened is mysterious. 2 
Abu Lubabah must have committed some grave fault not men- 
tioned in our sources. Probably he did not repudiate the old alliance 
of his clan ('Amr b. f Awf) with Qurayzah, but used his influence 
somehow or other in their favour. 

After the unconditional surrender of Qurayzah, Muhammad b. 
Maslamah was in charge of the men and 'Abdallah b. Sallam of the 
women and children. Some of the Aws are said to have appealed 
to Muhammad to forgive Qurayzah for the sake Of the Aws as he 
had pardoned Qaynuqa* for the sake of Ibn Ubayy and the Khazraj. 
Those who made this approach are not named, but subsequently 
four persons are said to have been gravely concerned at' the fate of 
Qurayzah, namely, ad-Dahhak b. Khallfah and Salamah b. Sala- 
mah (both of 'Abd al-Ash'hal), Mu'attib b. Qushayr (Dubay'ah 
of 'Amr b. 'Awf), and Hatib b. Umayyah (Zafar). This seems to 
indicate a wide-spread tendency in the Aws to honour the old 
alliance with Qurayzah. Muhammad met their request by suggest- 
ing that the fate of the Jews should be decided by one of their 
confederates, and to this they agreed. Muhammad therefore ap- 
pointed as judge Sa'd b. Mu'adh, the leading man of the Aws, who 
had been gravely wounded during the siege and died soon after his 
sentence on Qurayzah. When he was brought to where Muhammad 
was, all the Aws and the others present swore to abide by his 
decision. He decreed that all the men of Qurayzah should be put 
to death and the women and children sold as slaves. This sentence 
was duly carried out, apparently on the following day. 

1 IH, 684-99; WW, 210-24; Tab. 1485-98. * Cf. p. 188 above. 


Some European writers have criticized this sentence for what 
they call its savage and inhuman character. The general question 
involved will be dealt with later. 1 Here it is to be noticed that the 
participants in the events (and likewise the transmitters of the 
material) do not seem to have been concerned with the alleged 
harshness of the sentence. 2 The point at issue was whether allegi- 
ance to the Islamic community was to be set above and before all 
other alliances and attachments. In this connexion it must be 
remembered that the old Arab tradition was that you supported 
your confederate whatever his conduct towards other people might 
be, provided only that he remained faithful to you. It would seem 
then that those of the Aws who wanted leniency for Qurayzah 
regarded them as having been unfaithful not to the Aws but only 
to Muhammad ; that, means that they still regarded themselves as 
being primarily members of the Aws (or of some subdivision of it) 
and not of the Islamic community. It is thus unnecessary to sup- 
pose that Muhammad brought any pressure to bear on Sa'd b. 
Mu'adh to punish Qurayzah as he did. A far-sighted man like Sa'd 
must have realized that to allow tribal or clan allegiance to come 
before Islamic allegiance would lead to a renewal of the fratricidal 
strife from which they hoped the coming of Muhammad had saved 
Medina. As he was being led to the presence of Muhammad to 
pronounce his sentence, he is said to have remarked to those 
urging him to remember the old alliance that 'the time has come 
for Sa'd that no one's blame should touch him in respect of God', 
presumably meaning that, in view of the approach of death, he 
must perform his duty towards God and set the Islamic com- 
munity above the old confederacy; and it is noteworthy that the 
phrase rendered 'no one's blame' (lawmat loLim) occurs in a verse 
of the Qur'an warning the believers against 'drawing back' from 
their religion. 3 

We are given a glimpse of the potential dangers of the situation 
by the report of al-Waqidi that Sa'd b. 'Ubadah and al-Hubab b. 
al-Mundhir, leaders of the Khazraj, remarked to Muhammad that 
the Aws were not in agreement with the execution of the men of 

1 Cf. p. 328 below. 

2 Cf. 'The Condemnation of the Jews of Banu Qurayzah', Muslim World, 
xlii (1952), 160-71, esp. 171. (N.B. The second paragraph on p. 160 is entirely 
a quotation from Caetani.) 

3 IH, 689. i; Q. 5. 54/59. Cf. IS, iii/2. 4. 8, 'the time has come for me that 
no one's blame matters to me in respect of God'; also WW, 215. 


Qurayzah. This served, however, to put the Aws on their mettle, 
and Sa'd b. Mu'adh assured Muhammad that all the devout be- 
lievers among the Aws concurred in it. Thereupon two of the con- 
demned were given to each of the clans or sub-clans involved 
(' Abd al-Ash'hal, Harithah, Zafar, Mu'awiyah, 'Amr b. c Awf, and 
Umayyah b. Zayd), and these were duly executed, so that all the 
clans were involved in the blood of Qurayzah. As the execution is 
said to have been organized by 'All and az-Zubayr, the majority 
of the Jews (said to have numbered 600) were probably killed by 
Emigrants, though the Khazraj may also have helped, since apart 
perhaps from Ibn Ubayy they had no longer any alliance with 
Qurayzah. In the division of the palms there is no mention of any 
being given to B. c Awf of the Khazraj, to which Ibn Ubayy be- 
longed. 1 On the other hand, Sa'd b. 'Ubadah certainly took part 
in the affair, and his clan of Sa'idah is also omitted from the 
division of palms; this may be an error in the recorded list, but it 
may also be due to Sa'idah having numerous palms already Sa'd 
b. 'Ubadah had sometimes supplied the whole of Muhammad's 
forces with dates. 2 If the latter alternative is the true one, the 
absence of 'Awf and also Bayadah from the list of recipients of 
palms may be because they did not take any part in the fighting. 
Whatever the truth of the last detail, there must still have been 
much vigour in clan attachments and in the old ideas connected 
with them. The appointment of Sa'd b. Mu'adh as judge over 
Qurayzah was not an attempt by Muhammad to Conceal his 
alleged dictatorial power, since in fact at this period he had none; 
it was the only tactful way open to him of dealing with a difficult 

After the elimination of Qurayzah there remained no important 
group of Jews in Medina. There were still some Jews there, how- 
ever, and perhaps quite a number. One such was Abu 'sh-Shahm, 
who was attached to B. Zafar; he was a merchant and money- 
lender, and even bought some of the women and children of 
Qurayzah! 3 If the view of the dating of the Constitution to be 
propounded in the next chapter is sound, there must have been 
several small groups of Jews scattered about Medina. 

The continuing presence of at least a few Jews in Medina is an 

1 WW, 220. 

2 WW, 150, 163, 212; cf. 189, gift of a palm garden. 

3 IS, i/2. 173. 14; WW, 221; cf. WW, 174, 264 f., 278. 


argument against the view sometimes put forward by European 
scholars that in the second year after the Hijrah Muhammad 
adopted a policy of clearing all Jews out of Medina just because 
they were Jews, and that he carried out this policy with ever- 
increasing severity. In general it was not Muhammad's way to have 
definite policies of such a kind. What he did have was a balanced 
view of the fundamentals of the contemporary situation and of his 
long-term aims, and in the light of this he moulded his day-to-day 
plans in accordance with the changing factors in current events. 
The occasions of the attacks on Qaynuqa' and an-Nadir are no 
more than occasions (though they may well be genuine), and the 
historian is justified in looking for deeper underlying reasons. 
These are not far to seek. In Muhammad's first two years at Medina 
the Jews were the most dangerous critics of his claim to be a 
prophet, and the religious fervour of his followers, on which so 
much depended, was liable to be greatly reduced unless Jewish 
criticisms could be silenced or rendered impotent. It was difficult, 
however, for Muhammad to share with the rank and file of his 
followers his own appreciation of the importance of the Jews in the 
total religious-cum-political situation. When circumstances in 
general were favourable (e.g. when his own prestige was high and 
Ibn Ubayy's low) and an occasion of hostilities presented itself of 
the type familiar to the Arabs, then Muhammad acted. In a sense* 
therefore, his actions were spontaneous and not premeditated. 
Moreover, in so far as the Jews changed their attitude and ceased 
to be actively hostile, they were unmolested, as the case of Abu 
'sh-Shahm indicates. After the first incidents we may suppose that 
verbal criticisms of the Qur'anic revelation ceased except in strict 
privacy. This was replaced, however, by another form of hostile 
activity, diplomatic intrigue against Muhammad. With the fate of 
Qurayzah before their eyes the remaining Jews of Medina were 
presumably very circumspect and avoided all compromising rela- 
tionships, though at the time of the expedition to Khaybar their 
sympathies seem naturally to have been with their co-religionists. x 
Though the Jews of Medina had become quiescent, those at 
Khaybar, among whom the leaders of an-Nadir were the most 
prominent, were still anxious to avenge themselves on Muhammad. 
They made lavish, though no doubt judicious, use of their wealth 
to induce the neighbouring Arabs and especially the strong tribe 

1 Cf. WW, 264, also 266; contrast 283, Jews with Muhammad. 


of Ghatafan to join them against the Muslims. Muhammad had 
thus a straightforward reason for attacking Khaybar. The moment 
he chose for the attack May /June 628 (i/y) shortly after his return 
from the expedition of al-Hudaybiyah was one when it was also 
convenient for him to have booty to distribute to his followers 
whose expectations had recently been disappointed. The people 
of Khaybar had had some word of Muhammad's preparations, but 
his march to Khaybar was executed swiftly and secretly and they 
were taken by surprise with inadequate dispositions to resist a 
siege. Khaybar comprised several groups of strongholds, many 
built on the tops of hills and virtually impregnable. The Muslims 
attacked them piecemeal, beginning with the group known as 
an-Natat. There was much shooting from a distance and apparently 
some single combats. When the besieged made a sally, the Muslims 
fought back vigorously, and on at least one occasion followed them 
inside the gates. Several of the Muslim successes, however, were 
due to help they received from Jews who wanted in this way to 
ensure the safety of themselves and their families. When the strong- 
holds of an-Natat and those of ash-Shiqq had fallen there was little 
further resistance, and terms of surrender were speedily arranged 
for the remaining groups of strongholds, al-Katibah, al-Watih, and 
Sulalim. The principle was adopted that the Jews' should continue 
to cultivate the land, but should hand over half the produce to the 
Muslim owners the 1,600 participants in the expedition, or those 
to whom they had sold their shares. Several of the prominent men 
of Khaybar had been killed in single combat, and Kinanah b. Abi 
'1-Huqayq, apparently the chief leader, together with a brother, was 
put to death after the surrender because he had concealed the 
family treasure. Khaybar was thus reduced to a position of sub- 
servience and rendered innocuous. 1 

About the same time treaties were forced upon the colonies of 
Jews at Fadak, Wadi '1-Qura, and Tayma'. After the news of the fall 
of even a few of the strongholds of Khaybar there was no will to 
resist. The two former received similar terms to the men of Khay- 
bar, but the impost on the latter is called jizyah. It may be that 
they were treated differently because the two former had been 
actively hostile to Muhammad and had stirred up the neighbouring 
Arab tribes of Sa'd and possibly Badr b. Fazarah against him. 2 

1 IH, 756-81; WW, 264-96; Tab. 1575-90. 

2 Fadak: IH, 975; WW, 237 f. Wadi '1-Qura: IH, 979 f.; WW, 236 n. 


Various factors contributed to this Muslim success. The Jews 
were over-confident in the strength of their positions in Khaybar, 
and failed to lay in supplies of water sufficient for even a short 
siege. Man for man the Muslims were the better fighters, but this 
did not count for much in a siege except in so far as the besieged 
were forced to leave their strongholds through lack of water or 
other supplies. The Muslims seem to have been short of food for 
a time until they captured one of the strongholds with ample pro- 
visions. The lack of fundamental unity among the Jews was a weak- 
ness which meant that it was easy for Muhammad to find Jews 
ready to help him. Moreover, the Arab allies of the Jews were 
attached to them chiefly by bribes, and were therefore easily de- 
tached, partly by fear of Muslim reprisals and partly by Muham- 
mad's diplomatic skill. At Fadak B. Sa'd had been raided by 'All 
some months earlier, and Ghatafan, despite a show of support, 
made no effective intervention during the operations at Khaybar. 


The fall of Khaybar and surrender of the other Jewish colonies 
may be said to mark the end of the Jewish question during Muham- 
mad's lifetime, and this is not the place to discuss the expulsion of 
the Jews from the Hijaz by the caliph 'Umar. The Jews had 
opposed Muhammad to the utmost of their ability, and they had 
been utterly crushed. Many of them still remained in their former 
homes in Medina and elsewhere, but they had ceased to count in 
Arabian politics, and had lost much of their wealth. 

It is interesting to speculate on what would have happened had 
the Jews come to terms with Muhammad instead of opposing him. 
At certain periods they could have secured very favourable terms 
from him, including religious autonomy, and on that basis the 
Jews might have become partners in the Arab empire and Islam 
a sect of Jewry. How different the face of the world would be now, 
had that happened! In the early months at Medina the seeds 
were sown of a great tragedy; a great opportunity was lost. On 
the purely theological issues there would appear to be fewer 
difficulties in Islam for Jews than for Christians. But Muhammad's 
claim to receive messages from God conflicted with the cherished 
belief that the Jews were the chosen people through whom alone 
God revealed Himself to men. It was altogether in keeping with 
the traditional outlook of Jewry that the Jews of Medina should 


reject Muhammad. Even men more far-sighted than their actual 
leaders would have acted similarly. It was perhaps not necessary 
for the Jews to indulge in mocking criticism of Islam as they did; 
but, once they had decided to reject Muhammad, they had to 
justify this action at least to themselves. Their criticism was a threat 
to the whole social and political experiment in which he was 
engaged, and could not be ignored. Thus the whole sorry train of 
events was set in motion. 

To suggest that Muhammad was unaware of the wealth of the 
Jews would be a serious underestimate of his intelligence. To make 
this the sole reason, however, for his attacks on the Jews is to be 
unduly materialistic. The wealth of the Jews was certainly of great 
benefit to him and considerably eased his financial position, and 
the prospect of financial betterment may have 'Influenced the timing 
of his attacks on the Jews. But the fundamental reason for the 
quarrel was theological on both sides. The Jews believed that God 
had chosen them specially, Muhammad realized that his prophet- 
hood was the only possible basis of Arab unity. As so often in the 
history of the Middle East, theology and politics were intermingled. 



IBN ISHAQ has preserved an ancient document commonly known 
as the 'Constitution of Medina*. Apart from the introductory 
words, however, he tells us nothing about it, neither how he 
came by it nor when and how it was brought into force. On the 
latter points he must be presumed ignorant; its place near the 
beginning of his account of the Medinan period is simply that 
called for by logic. 

(a) The text of the document 1 

Ibn Ishaq said: The Messenger of God (God bless and preserve 
him) wrote a writing (kitdb) between the Emigrants and the Ansar, 
in which he made a treaty and covenant with the Jews, confirmed 
them in their religion and possessions, and gave them certain 
duties and rights : 

In the name of God, the Merciful, the Compassionate! 

This is a writing of Muhammad the prophet between the be- 
lievers and Muslims of Quraysh and Yathrib and those who follow 
them and are attached to them and who crusade (jdhadu) along 
with them. 

1. They are a single community (ummak) distinct from (other) 
people. 2 

2. The Emigrants of Quraysh, according to their former con- 
dition, 3 pay jointly the blood-money between them, and they (as 
a group) ransom their captive(s), (doing so) with uprightness and 
justice between the believers. 

1 IH, 341-4. The numbering of the paragraphs follows Wensinck, Mohammed 
en dejoden, 74-81, except that the closing sentence of 19 has been moved there 
from the beginning of 20. Cf. also Wellhausen, Skizzen, iv. 65-83, and Caetani, 
i. 391-408. 

2 The literal translation of the last phrase is 'from the people', which might 
refer to the Jews ; but on the whole this is unlikely. 

3 Lane, s.v., makes it clear that the phrase 'aid rib* ati-him means According 
to their former or good condition*. There is no reason to suppose any reference 
to 'quarter*. The interpretation is either that each group remains distinct or that 
it follows its previous practice. The last clause prescribes a fair apportionment 
between the various groups within the clan. 


3. Banu 'Awf, according to their former condition, pay jointly 
the previous blood-wits, and each sub-clan (td'ifah) ransoms its 
captive(s), (doing so) with uprightness and justice between the 
believers. 1 

4. Banu '1-Harith, according to their former condition, pay 
jointly ... (as 3). 

5. Banu Sa'idah ... (as 3). 

6. Banu Jusham ... (as 3). 

7. Banu 'n-Najjar ... (as 3). 

8. Banu 'Amr b. 'Awf ... (as 3). 

9. Banu 'n-Nabit ... (as 3). 

10. Banu '1-Aws ... (as 3). 

11. The believers do not forsake a debtor among them, but give 
him (help), according to what is fair, for ransom or blood-wit. 

12. A believer does not take as confederate (hallf) the client 
(mawla) of a believer without his (the latter's) consent. 

13. The God-fearing believers are against whoever of them acts 
wrongfully or seeks ( ? plans) an act that is unjust or treacherous 
or hostile or corrupt among the believers ; their hands are all against 
him, even if he is the son of one of them. 

14. A believer does not kill a believer because of an unbeliever, 
and does not help an unbeliever against a believer. 

15. The security (dhimmah) of God is one; the granting of 
'neighbourly protection' (yujlr) by the least of them (the believers) 
is binding on them ; the believers are patrons (or clients mawdli) 
of one another to the exclusion of (other) people. 

1 6. Whoever of the Jews follows us has the (same) help and 
support (nasr, iswah) (as the believers), so long as they are not 
wronged (by him) and he does not help (others) against them. 

17. The peace (silni) of the believers is one; no believer makes 
peace apart from another believer, where there is fighting in the 
way of God, except in so far as equality and justice between them 
(is maintained). 

1 8. In every expedition made with us the parties take turns with 
one another. 2 

19. The believers exact vengeance for one another where a man 

1 'The previous blood-wits' (al-ma'dqil al-uld) are those according to the 
principles previously in force. The words 'between the believers' may be in- 
tended to exclude unbelievers belonging to B. 'Awf. 

2 This may apply to taking turns at riding a camel (Wellhausen; cf. IH, 433, 
&c.), or to all military duties (Caetani). 


gives his blood in the way of God. The God-fearing believers are 
under the best and most correct guidance. 

20. No idolater (mushrik) gives 'neighbourly protection' (yujir) 
for goods or person to Quraysh, nor intervenes in his (a Qurashi's) 
favour against a believer. 

21. When anyone wrongfully kills a believer, the evidence being 
clear, then he is liable to be killed in retaliation for him, unless the 
representative of the murdered man is satisfied (with a payment). 
The believers are against him (the murderer) entirely; nothing is 
permissible to them except to oppose him. 

22. It is not permissible for a believer who has agreed to what 
is in this document (sahlfah) and believed in God and the last day 
to help a wrong-doer 1 or give him lodging. If anyone helps him 
or gives him lodging, then upon this man is the curse of God and 
His wrath on the day of resurrection, and from him nothing will 
be accepted to make up for it or take its place. 

23. Wherever there is anything about which you differ, it 
is to be referred to God and to Muhammad (peace be upon 

24. The Jews bear expenses along with the believers so long as 
they continue at war. 

25. The Jews of Banu f Awf are a community (ummah) along 
with the believers. To the Jews their religion (din) and to the 
Muslims their religion. (This applies) both to their clients and to 
themselves, with the exception of anyone who has done wrong or 
acted treacherously; he brings evil only on himself and on his 

26. For the Jews of Banu 'n-Najjar the like of what is for the 
Jews of Banu ' Awf. 

27. For the Jews of Banu '1-Harith the like . . . 

28. For the Jews of Banu Sa'idah the like . . . 

29. For the Jews of Banu Jusham the like . . . 

30. For the Jews of Banu '1-Aws the like . . . 

31. For the Jews of Banu Tha'labah the like of what is for the 
Jews of Banu 'Awf, with the exception of anyone who has done 
wrong or acted treacherously ; he brings evil only on himself and 
his household. 

32. Jafnah, a subdivision (batn) of Tha'labah, are like them. 

1 Muhdith, literally 'innovator', means one who disturbs the existing state of 
affairs in any way. 


33. For Banu 'sh-Shutaybah 1 the like of what is for the Jews of 
Banu 'Awf ; honourable dealing (comes) before treachery. 2 

34. The clients of Tha'labah are like them. 

35. The bifdnah 3 of (particular) Jews are as themselves. 

36. No one of them ( ? those belonging to the ummah) may go 
out (to war) without the permission of Muhammad (peace be upon 
him), but he is not restrained from taking vengeance for wounds. 
Whoever acts rashly (fataka), it (involves) only himself and his 
household, except where a man has been wronged. God is the 
truest (fulfiller) of this (document). 4 

37. It is for the Jews to bear their expenses and for the Muslims 
to bear their expenses. Between them (that is, to one another) 
there is help (nasr) against whoever wars against the people of this 
document. Between them is sincere friendship (nash wa-nasihah), 
and honourable dealing, not treachery. A man is not guilty of 
treachery through (the act of) his confederate. There is help for 
(or, help is to be given to) the person wronged. 

38. The Jews bear expenses along with the believers so long as 
they continue at war. 

39. The valley of Yathrib is sacred for the people of this 

40. The 'protected neighbour' (jar) is as the man himself so long 
as he does no harm and does not act treacherously. 

41. No woman is given 'neighbourly protection' (tujdr) without 
the consent of her people. 

42. Whenever among the people of this document there occurs 
any incident (disturbance) or quarrel from which disaster for it 
(the people) is to be feared, it is to be referred to God and to 
Muhammad, the Messenger of God (God bless and preserve him). 
God is the most scrupulous and truest (fulfiller) of what is in this 

43. No 'neighbourly protection* is given (Id tujdr) to Quraysh 
and those who help them. 

1 Wensinck, Jo den, 79, corrects to Banu 'sh-Shutbah; cf. as-Samhudi, 151. 

2 Or 'honourable dealing without treachery (is demanded)'. 

3 The meaning of bitdnah is obscure. It probably means those who were 
closely connected with some Medinan Jews by ties of friendship, not of blood ; 
cf. Q. 3. 118/114; IH, 519. 4; Aghani, xvii. 56. 22. Wensinck, 78, with some 
likelihood thinks they may be those Arabs who had been associated with the 
Jews before the coming of the Aws and the Khazraj. 

4 The second half of this article, and especially the last sentence are uncertain 
in meaning. The last sentence might mean 'God is very far from this.' 


44. Between them ( ? the people of this document) is help 
against whoever suddenly attacks Yathrib. 

45. Whenever they are summoned to conclude and accept a 
treaty, they conclude and accept it; when they in turn summon to 
the like of that, it is for them upon the believers, 1 except whoever 
wars about religion; for (? = incumbent on) each man is his share 
from their side which is towards them. 

46. The Jews of al-Aws, both their clients and themselves, are 
in the same position as belongs to the people of this document 
while they are thoroughly honourable in their dealings with the 
people of this document. Honourable dealing (comes) before 

47. A person acquiring ( ? guilt) 2 acquires it only against him- 
self. God is the most upright and truest (fulfiller) of what is in this 
document. This writing does not intervene to protect a wrong-doer 
or traitor. He who goes out is safe, and he who sits still is safe in 
Medina, except whoever does wrong and acts treacherously. God 
is 'protecting neighbour' (jar) of him who acts honourably and 
fears God, and Muhammad is the Messenger of God (God bless 
and preserve him). 

(b) The authenticity, date, and unity of the document 

This document has generally been regarded as authentic, though 
it has not always been given the prominence appropriate to an 
authentic document of this sort. The reasons for its authenticity 
have been succinctly stated by Wellhausen. 3 No later falsifier, 
writing under the Umayyads or 'Abbasids, would have included 
non-Muslims in the ummah, would have retained the articles 
against Quraysh, and would have given Muhammad so insignificant 
a place. Moreover the style is archaic, and certain points, such as 
the use of 'believers' instead of 'Muslims' in most articles, belong 
to the earlier Medinan period. 

There has been some discussion, however, whether the docu- 
ment is to be dated before or after the battle of Badr. Wellhausen 
placed it before Badr. Hubert Grimme, 4 however, argued for a date 

1 This may mean 'it is a debt owed to them by the believers' (cf. W. Wright, 
Arabic Grammar 3 , Cambridge, 1896-8, ii. 169 a), or 'it is for them to conclude 
without taking notice of the believers' (cf. ibid. 172 a). The interpretation of this 
article is obscure. 2 Cf. Q. 4. in, cited by Wensinck. 

3 Skizzen, iv. 80; cf. Caetani, i. 403. 

4 Muhammed, Munster, 1892, i. 76. 
6783 Q 


after Badr on the following grounds : the functions attributed to 
Muhammad in 23 and 36 show that his authority was generally 
recognized; the references to fighting for the faith (ft sabil Allah, 
17, i<)',fi 'd-dtiiy 45) imply that some fighting had taken place; 
the hostile attitude towards Quraysh could have been demanded 
of Medinan believers only after Badr. Caetani 1 shows that these 
arguments are not so strong as Grimme thought, and prefers a date 
prior to Badr. 

This discussion of the date has assumed that the document is 
a unity; but that is the point that ought to be examined first. There 
are reasons for thinking that articles which originated at different 
dates have been collected. 2 Thus there are certain linguistic varia- 
tions : the believers are mostly spoken of in the third person, but 
sometimes they are 'you' and sometimes 'we' (as in 23, 16, 18); 
mostly they are 'believers', but twice they are 'Muslims' ( 25, 37). 
Again, certain articles come near to being repetitions of other 
articles; they deal with the same problem but may have slight 
alterations. Both 23 and 42 say that disputes are to be referred 
to Muhammad, though 42 is more precise. Both 20 and 43 are 
directed against Quraysh. The points about Jews in 16 and 24 
are similar to those in 37 and 38; and indeed 24 and 38 
are identical. Finally both 30 and 46 deal with the Jews of the 
Aws. It is to be noted that the articles which are similar do not 
occur together, as one would expect where articles dealt with 
different aspects of the same point. On the contrary one set is 
spread between 16 and 30 and another set between 37 and 46. 
This is sufficient to justify an examination of the possibility that 
the document as we have it contains articles from two or more 
different dates. 

With this possibility in mind let us turn to what is said about 
the Jews. The inclusion of the Jews in the ummah is an important 
argument for dating the document before Badr. 3 The omission of 
the names of the three great Jewish tribes or clans is surprising. 
One way of explaining it, however, is to suppose that Muhammad 
grouped the Jews according to the Arab clans in whose districts 
they lived; an- Nadir and Qurayzah would then be included among 
the Jews of al-Aws and Tha'labah, since they lived between 

1 Op. cit. 404. 

2 I am here indebted to the late Richard Bell, who, by his insistence on this 
point in conversation, led me to examine it carefully. 

3 Cf. Wellhausen, ibid. 80. 


Awsallah and Tha'labah b. c Amr b. 'Awf. 1 There are strong 
reasons, however, for thinking that the three main Jewish groups 
are not included in the document. For one thing it is most likely 
that a phrase like 'the Jews of the Banu 'Awf * means the Jews who 
were confederates of that clan. Small groups of Jews, like those at 
Ratij, 2 doubtless became confederates of the Arab clan surrounding 
them ; but an-Nadir and Qurayzah had their own territories, and 
were latterly confederates of 'Abd al-Ash'hal, who lived some 
distance away, and who were part of the clan of an-Nabit which is 
not mentioned in 25-35 among the clans with Jews attached. 
Secondly, Ibn Ishaq 3 has a list of sixty-seven Jewish opponents of 
Muhammad and arranges them under the following heads : B. an- 
Nadir (12), B. Tha'labah b. Fityawn (3), B. Qaynuqa' (31), B. 
Qurayzah (17), JevCs of B. Zurayq (i), Jews of B. Harithah (i), 
Jews of B. 'Amr b. 'Awf (i), Jews of B. an-Najjar (i). This makes 
it probable that 'the Jews of B. Tha'labah' of 31 are those whom 
Ibn Ishaq and as-Samhudi 4 reckon as a Jewish clan, and shows 
that at some period small groups of Jews, distinct from the three 
main clans, were known as 'the Jews of such-and-such an Arab 

It seems probable, then, that the three main Jewish groups are 
not mentioned in the document. If that is so, the document in its 
present form might belong to the period after the elimination of 
Qurayzah. The difficulty that much attention is given to Jewish 
affairs at a time when there were few Jews in Medina could be 
explained by the hypothesis that the document in its final form 
was intended as a charter for the Jews remaining in Medina and 
included all relevant articles from earlier forms of the Constitution 
of the city. 

The history of the document might be reconstructed conjectur- 
ally somewhat as follows. The earlier articles (up to 15 or 1 6 or 
19 or 23) may have been the original terms of agreement between 
Muhammad and the Medinan clans at al-'Aqabah, or they may 
have been drawn up by the 'representatives' (nuqabff) shortly after 
the Hijrah. They mostly deal with problems involved in keeping 
peace between the Arab clans. To these from time to time as need 
arose other articles were added, while articles which became in- 
operative would be dropped, e.g., articles about Qurayzah and 

1 Cf. ibid. 75. 2 Cf. above, pp. 160, 194. 

3 IH, 351 f- 4 i- "5- 


an-Nadlr. The word sahifah (translated 'document'), which occurs 
from 22 to 47 implies a written document formally accepted by 
different parties. The phrase 'the people of this document* is doubt- 
less used so as to cover both Jews and Muslims. To the 'document' 
in this special sense belongs the solid body of articles dealing with 
Jews, 24 to 35 (or, if 36 is interpreted as referring to the Jews, 
to 38). 1 6 is perhaps part of the 'Aqabah agreement with the 
Aws and the Khazraj, and prior to the formal agreement with the 
Jews in the sahifah or 'document*. 

While scholars may come to approve some such view of the 
existing text of the Constitution of Medina, there is much that is 
bound to remain conjectural and obscure. Thus, is 44 an earlier 
version of the middle clause of 37? Are the Jews of Band 'Awf 
given a special place because 'Abdallah b. Ubayy first obtained 
good terms for them ? Why are the Jews of Banu '1-Aws mentioned 
twice? Is Banu J l-Aws here and in 30 identical with Banu '1-Aws 
in 10 (which is commonly taken to be the group usually known 
as Awsallah), or is it the whole tribe of the Aws ? This is not the 
place to pursue such queries further. This study of the text of the 
Constitution, however, is sufficient to justify the use of it as a source 
for the ideas underlying the Islamic state in the early formative 
years, while at the same time it warns us not to base an argument 
solely on the supposed date of any article of the Constitution. 


The Constitution of Medina is not certain evidence of the 
position taken by Muhammad in Medina when he arrived there 
in September 622 (iii/i), but his powers under the Constitution 
are so slight that they cannot have been much less at the beginning 
of his residence in Medina. All that the Constitution explicitly 
states is that disputes are to be referred to Muhammad ( 23, 42). 
In addition the phrase 'Muhammad the prophet' occurs in the 
preamble; and the appearance of the Muhajirun or Emigrants on 
the same level as one of the Medinan clans implies that Muhammad 
as chief of the Emigrants was on a level with the chiefs of the 
various clans. As the Emigrants are mentioned first, perhaps Mu- 
hammad had a primacy of honour among the chiefs of the clans. 
He is very far, however, from being autocratic ruler of Medina. 
He is merely one among a number of important men. During his 
first year in Medina several others were probably more influential 


than Muhammad. The provision that disputes were to be referred 
to him would not in itself increase his power, unless he had 
sufficient tact and diplomacy to find a settlement that would com- 
mand general agreement. 

Various incidents of the first half of the Medinan period show 
the theoretical weakness of Muhammad's position. After the 
* affair of the lie' against 'A'ishah's chastity, in which Ibn Ubayy 
had been active in spreading the calumny, Muhammad could not 
take direct action against him, but had to call a meeting of the 
Ansar and ask permission of those who might have felt that it was 
obligatory for them to avenge any injury to Ibn Ubayy. In this case 
Muhammad easily gained his point, for, whether by design or 
accident, the enmity of the Aws and the Khazraj was fanned into 
flame, and the great decline in Ibn Ubayy's influence became 
apparent. 1 Similarly, when the question of punishing B. Qurayzah 
for their disloyalty arose, Muhammad did not venture to pronounce 
any judgement himself, since, had he decreed any shedding of 
blood, honour might have impelled some confederates of B. Quray- 
zah to avenge it, even though they were Muslims. The decision 
about the punishment was left to the chief of the clan of which they 
had been confederates. 2 

These are clear examples of how the Medinan clan-chiefs re- 
tained much of their power and thereby limited Muhammad's 
authority. They are not isolated examples, however. The whole 
story of his physical attacks on Jews presupposes the old back- 
ground of clan-relationships, and shows how these had always to 
be considered in choosing agents. 3 Muhammad is seen to be the 
chief of one of several co-operating groups, with little to mark him 
out from the others. 

The referring of disputes to Muhammad is closely connected 
with the recognition of him as prophet. The wording of the Consti- 
tution is that disputes are to be referred to God and to Muhammad. 
The idea that one of the functions of a prophet is to mete out 
justice occurs in a Meccan passage of the Qur'an : 'each community 
has a messenger, and when their messenger comes, judgement is 
given between them with justice, and they are not wronged'. 4 This 
point was doubtless realized by the Medinans when they recognized 
Muhammad as prophet; part of what attracted them to him was 

1 Cf. above, p. 186. 2 Cf. above, p. 214. 

3 Cf. above, pp. 181 f., &c. 4 10. 47/480. 


the hope that he would be able to put an end to the internal dis- 
putes that made life in Medina intolerable. 1 If the Medinans did 
not explicitly admit this right to judge disputes when they acknow- 
ledged Muhammad as prophet and arranged for him to come to their 
city, they must soon have been forced to do so, since a revelation 
came commanding the reference of disputes to God. 2 This should 
properly mean that disputes were to be settled by a specific revela- 
tion from God to Muhammad; but doubtless in practice Muham- 
mad was held to have the best knowledge of what God's decision 
would be on a case where there was no specific revelation. The 
opening words of this verse, * wherever there is anything about 
which you differ*, are identical with those of 23 of the Constitu- 
tion ; this suggests that the connexion is close. The phrase is a vague 
one and could be applied to far-reaching differences on policy as 
well as to petty quarrels between neighbours. 

Such, then, is the position of Muhammad as stated in the 
Constitution and portrayed in the history of his early years in 
Medina. How far this position was agreed upon in the meetings 
at al-'Aqabah it is impossible to say. The ' pledge of the women' 
{bay' at an-nisd y ) follows the text of a verse of the Qur'an revealed 
after al-Hudaybiyah, and cannot be accepted as evidence of the 
content of a promise made to Muhammad. 3 According to the 
traditional account of the second meeting of al-' Aqabah, a further 
pledge was made there, known as the 'pledge of war' (bay'at 
al-harb). In Ibn Ishaq's version the important words are those 
spoken by al-Bara' b. Ma'rur, 'By Him who sent thee a prophet 
with the truth, we shall defend thee from that from which we 
defend ourselves' (or 'our wives and families'). 4 Apart from the 
reference to Muhammad's prophethood, there is nothing in these 
words to suggest that this alliance between Muhammad and the 
Medinans is different from any other alliance. The same holds of 
Muhammad's reply to a question about the possibility of his 
receiving a revelation commanding him to return to Mecca; he 
said he regarded himself and the Medinans as belonging to one 
another, and that he would fight those against whom they fought, 
and make peace with those with whom they made peace. This is 

1 Cf. T. Noldeke, Geschichte des Qordns, ed. F. Schwally, Leipzig, 1909, i. 165. 

2 42. IO/SE; cf. 4. 59/620; 24. 47/46-52/51 G may refer to a later attempt 
to reverse the argument. 

3 60. 12 G; IH, 289; cf. MjMecca, 146 ; G. H. Stern in Bulletin of the School 
of Oriental Studies, x. 185-97. 4 IH, 296. 


just a military alliance. The accounts tell us nothing about Mu- 
hammad's position in the Medinan polity, apart from the fact that 
he was acknowledged as prophet; and this we would in any case 
have presumed. It may well be that, until he went to Medina, 
Muhammad was content with the recognition of his prophethood 
and asked for no further privileges (unless something is implicit 
in his request for nuqabff or representatives to confer with him). 
In so far as his prophethood was recognized he would have a 
starting-point from which he could begin to build up his power. 1 

The most mysterious aspect of Muhammad's position when he 
went to Medina is the military one. The words of the * pledge of 
war* speak of defensive action only; they say nothing about offen- 
sive operations, and even in the case of defence they do not say any- 
thing about who was to lead. What happened was that the first expe- 
ditions were offensive expeditions from Medina in the hope of 
ambushing a Meccan caravan. It is not certain that the Medinans 
took part in these expeditions, but the probability is that they did. 2 
In every case the leader was either Muhammad himself or one of 
the Emigrants appointed by him. This was doubtless not because 
of any unrecorded agreement that the Emigrants were to lead in 
war, but because the organizers of these expeditions were the 
Emigrants, while the Ansar were merely invited to join. It is 
expressly stated that Muhammad called for volunteers for the 
expedition of 'Ushayrah. 3 As these expeditions, even that to Badr, 
were razzias, where the aim was to capture booty without undue 
danger to oneself, the Ansar presumably did not think that they 
would provoke a great expedition against Medina, such as that of 
the Meccans to Uhud. Muhammad seems to have done what he 
could to collect men for Badr, but apparently not even all who 
sincerely believed in his prophethood joined in; and we are told 
that those who did not join were not blamed. It must therefore 
have been by invitation and exhortation that Muhammad obtained 
his 300 or so men. 

The booty captured at Badr was apparently disposed of by 
Muhammad as he pleased, and this confirms the view that the 
expedition was, as it were, a private one organized by him, which 
he invited others to join. Before the battle he is said to have 
promised certain rewards to those who killed or captured an 

1 Cf, E. E. Evans-Prichard, The Sanusi of Cyrenaica, Oxford, 1949, 59 f. 

2 Cf. p. 3 above. 3 WW, 34. 


enemy; and apparently, after fulfilling these promises (and pre- 
sumably retaining some for his own use), the rest of the spoil was 
divided equally among the participants. Muhammad may subse- 
quently have felt, however, that this way both of fighting and of 
dividing the spoil was unsatisfactory; or the repercussions of Badr 
in Medina may have necessitated changes. At any rate, by the time 
of the expedition against Qaynuqa' in the month after Badr, it 
had been decreed that a fifth (khums) of all spoils taken on a Mus- 
lim expedition was to go to Muhammad. This change, moreover, 
implies several other changes. For one thing it implies that Mu- 
hammad had been recognized as in some sense chief of the ummah. 
It was customary in Arabia for the chief of a tribe to receive a 
quarter of the spoils, partly for his own use, but partly in order to 
perform certain functions on behalf of the tribe, such as looking 
after the poor and giving hospitality. 1 The change from a quarter 
to a fifth marks off the head of the ummah from tribal chiefs ; yet 
the verse prescribing the fifth 2 (which was perhaps revealed 
immediately after Badr) indicates that the fifth was in part to be 
used by Muhammad for these communal purposes. 

Such arrangements, again, together with the recognition of 
Muhammad as head of the ummah, show that he had managed, in 
the enthusiasm after the victory, to persuade most of the ummah 
to accept the consequences of Badr. All must have realized that 
the Meccans would try to avenge the bloodshed; and the more 
level-headed may have suspected that they might not defeat the 
Meccans so easily another time. Despite the anxious future, how- 
ever, the Ansar in general resolved to support Muhammad more 
fully. To strengthen their resolution there were revelations bidding 
them to fight the Meccans till ' there is no more persecution and 
the religion is entirely God's'. 3 This, too, is the most likely period 
for the inclusion in the Constitution of articles directed against 
Quraysh, and also of others emphasizing the unity of the ummah 
in war and peace. It is not easy to see how the Ansar could have 
been brought to accept such articles earlier while there was still no 
break with Mecca. 

As it was, there was some opposition in which Ibn Ubayy was no 

1 Abu Tammam, Hamdsah, ed. G. Freytag, Bonn, 1828-47, i. 458; Aghdni 
xvi. 50; Lane, s.v. rub'] Buhl, Mohammed, 31 n. 

2 8 41/42 F; cf. Lammens, Mecque, 153/249. 

3 8. 39/40 F; 47. 4 is probably later than Badr, if the fighting there was unpre- 


doubt prominent. The question asked by this party, 'Why has not 
a surah been sent down?', implies that they professed themselves 
ready to act on a revelation from God, but not on the mere word 
of Muhammad; but, when 'a clearly formulated surah* did come, 
their misgivings were not allayed. 1 The period between Badr and 
the siege of Medina must have been a difficult one for Muhammad, 
when he was endeavouring to establish his ascendancy in Medina. 1 
The words 'obey God and His messenger' and various equivalents 
occur about forty times in the Qur'an, and are to be dated mostly 
in the months before and after the battle of Uhud. There is a series 
of stories of earlier prophets where these are made to say to their 
hearers 'fear God and obey me'. 2 There are passages where those 
who obey God and His messenger are promised the delights of 
Paradise, 3 while those who have not obeyed repent at leisure in 
the Fire (that is, Hell). 4 Some passages refer to particular points 
(such as the disposal of spoils, 5 and the prohibition of wine and the 
game of maysir) 6 where there is no obvious connexion with circum- 
stances about the time of Uhud. A few seem to be later. 7 Many, 
however, refer to the opposition which Muhammad's policy 
encountered after Badr, either in general 8 or over some particular 
point, such as fighting the Meccans or bringing disputes to Mu- 
hammad. 9 This shows that it was becoming necessary for Muham- 
mad to insist on his own special position. He appears, however, 
to have regarded such passages as exhortations and not as com- 
mands. The Constitution does not prescribe obedience; and it is 
not for disobedience but for faintheartedness that Ibn Ubayy and 
his supporters are reproached after Uhud. 10 We must therefore 
conclude that at this period general obedience to Muhammad (as 
distinct from obedience to specific precepts of the Qur'an) was not 
formally prescribed. 

This state of affairs must be presumed to have continued at least 

1 Q. 47. 20/22, 33/35 ff.; cf. p. 183 above. 

2 26. 108-79, eight instances, added in Medina according to Bell; cf. 43. 63 E; 
7i. 3 E+; 3- 50/44 FG. 

3 4. 13/17 G, 69/71 G; 33. 71 ; 48. 16 f. Hi; 49. 14 HI. 

4 33- 66 f. E. 

5 8. i F, almost certainly revealed at Badr; for a discussion of the meaning of 
anfdl, cf. Tab., Tafsir, ix. 106-12. 6 5. 91/93 G. 

7 9. 71/72 ? HI; 48. 1 6 f. HI; 49. 14 HI; and perhaps others. 

8 24. 54/53 G; 3. 32/29 G; cf. 4. 80/82 G; &c. 

9 47- 33/35 ff- FG; 24. 47/46 fT. G; 4. 59/62 GH. 
10 Cf. p. 184 above. 


until the expedition of al-Hudaybiyah in March 628 (xi/6). The 
punishment of Abu Lubabah in May 627 (xii/5) was inflicted not 
by Muhammad but by himself; and he appears to have been 
released not by an order from Muhammad but by a revelation 
from God (though none of the suggested verses of the Qur'an fits 
the occasion). It may well be, therefore, that 'the pledge of good 
pleasure', as suggested above, was a pledge to do whatever 
Muhammad commanded, that is, to obey him. 1 Whether obedience 
was expected of all Muslims or only of those who pledged them- 
selves is not clear ; but, even if for a time those who did not pledge 
themselves were not formally bound to obey Muhammad, it would 
be increasingly difficult for them to oppose him. He was growing 
stronger, and, when weak tribes asked for alliance, was demanding 
a promise to obey. A Qur'anic verse denouncing those who oppose 
the decisions of Muhammad (33. 36) possibly belongs to the year 
628; most of the surah can be dated in 627 (5-6), but Richard Bell 
regards this verse as a later addition. 2 

A matter which might have thrown light on the extent of 
Muhammad's authority after al-Hudaybiyah is unfortunately 
obscure. He is said to have remarked to some men of the Medinan 
clan of Salimah, 'Who is your chief (sayyid)? When they replied 
'al-Jadd b. Qays J , Muhammad said, 'No, it is Bishr b. al-Bara' b. 
Ma'rur'. 3 This might be the formal deposition of a clan chief, but 
it might also be merely a hint to clansmen loyal to Muhammad 
that they ought to depose al-Jadd (or perhaps even just a compli- 
ment to Bishr). Though this seems a curious way to depose a chief 
formally, yet it is conceivable that it was something like a formal 
deposition. If so, it was an exercise of Muhammad's authority 
following on 'the pledge of good pleasure'. Indeed the incident is 
closely connected with the pledge, and probably occurred almost 
immediately afterwards. Al-Jadd was the one man in Muhammad's 
party at al-Hudaybiyah who refused to make the pledge, while 
Bishr was a suitable person for chief, being the son of al-Bara' b. 
Ma'rur, the chief of Salimah who first made 'the pledge of war' 
at al-'Aqabah but died about the time of the Hijrah. Bishr ate of 
the poisoned fish at Khaybar, two months after al-Hudaybiyah. 
All we can say, then, is that this may be an instance of Muhammad 
using his autocratic power and is then to be dated immediately 

1 Cf. p. 50 above. 2 Translation, ad he. 

3 IS, iii/2, ii2;cf. WW, 248. 


after the pledge at al-FIudaybiyah. If it is merely a hint to the loyal 
clansmen, it is an interesting example of how Muhammad obtained 
the decision he wanted in matters where he had no formal authority. 

An incident from early 631 (late 9) shows that by that time 
Muhammad was being obeyed. When Ka'b b. Malik and two 
other men stayed away from the expedition to Tabuk without any 
reasonable excuse, they were 'sent to Coventry' by the Muslims, 
and this was done by Muhammad's order. Even here, however, it 
is interesting to note that Muhammad insisted that the repeal of 
the sentence came not from himself but from God (probably in 
Qur'an 9. nS/iig). 1 The excommunication of the 'hypocrites' of 
this period was probably based on revelation and not on Muham- 
mad's order. 2 

The treaties and letters whose text is given by Ibn Sa'd (i/2) 
mostly come from the closing years of Muhammad's life. It is not 
surprising, then, to find Muhammad's name coupled with that of 
God in such phrases as 'the security-guarantee (dhimmaK) of God 
and of Muhammad b. 'Abdallah' and 'secure with the security 
(amin bi-amdn) of God and the security of Muhammad'. 3 The 
obligation to obey Muhammad is not mentioned except in a few 
documents; 4 but, though it is not mentioned, it is often implicit. 
The extant documents thus confirm the view that from a date not 
later than the conquest of Mecca Muhammad was acting as 
undisputed head of the Muslim community in political as well as 
religious matters. If Muhammad wrote letters to the heads of 
neighbouring states after al-Hudaybiyah (though the traditional 
account of the contents may be regarded as incorrect), that would 
suggest that it was about this time that he became conscious of 
having overcome all serious opposition. 5 It may also be that, in 
accordance with pre-Islamic custom, the use of titles like 'Mu- 
hammad the Prophet' and 'the Messenger of God' involved a claim 
to political leadership. 6 

The extent of Muhammad's autocratic powers in his last two 
or three years is further illustrated by his appointment of 'agents' 
to act on his behalf in various areas, and indeed by the whole 

1 IH, 907/13; WW, 411-14. 2 Cf. p. 190 above. 

3 IS, i/2. 23. 26 ( 25); 25. 2 ( 30 d); &c. 

4 Ibid. 23. 10 ( 23); 25. i ( 30 d); &c. 5 Cf. p. 41 above. 

6 Cf. J. Ryckmans, L* Institution Monarchique en Arabic Meridionale avant 
V Islam, 327-9; inscriptions show the importance of the royal epithet and the 
ceremony of taking it. 


matter of administrative appointments. From the beginning Mu- 
hammad had appointed men to perform various functions for 
which he was responsible. Thus he appointed commanders for the 
expeditions where he was not present in person. Until Badr he may 
have done so as head of the Emigrants, but latterly he was acting 
on behalf of the Muslim community as a whole, since some of the 
leaders were from the Ansar or even from nomadic tribes. 1 It may 
sometimes have happened that a man who had a private feud 
against an enemy of Muhammad's was allowed to organize an ex- 
pedition in the name of the community ; the leader thus recognized 
doubtless agreed to pay Muhammad the 'fifth' of the booty. 
'Uyaynah even seems to have led such an expedition against 
Tamlm before he was a Muslim. 2 Another regular appointment 
from the earliest times was that of a deptity in Medina when 
Muhammad was absent from the city. During the Badr expedition 
there was another deputy in the suburb of Quba', perhaps because 
this district was still mainly non-Muslim. 3 Other appointments 
from an early date were of men to supervise the booty and the 
prisoners. 4 All these were in spheres where Muhammad's right 
to order matters was accepted. The Constitution ( 36) states that 
warlike expeditions required Muhammad's permission. 

As Muhammad's influence expanded, further appointments 
were needed. Thus, even after the spoils of Khaybar had been 
divided out, it was necessary to have an inspector to estimate and 
receive half of the annual harvest. 5 We hear of governors of the 
neighbouring Jewish settlements of Wadi '1-Qura and Tayma'. 6 In 
his dealings with tribes in the neighbourhood of Medina Muham- 
mad made use of the leading men of the tribe in so far as these were 
friendly to him. One of the earliest functions entrusted to these men 
was that of summoning their fellow tribesmen for Muhammad's 
expeditions, such as those of the conquest of Mecca and of Tabuk ; 7 
those who summoned the tribes were also to a large extent their 

1 E.g. Muhammad b. Maslamah, Ghalib b. 'Abdallah al-Laythi; cf. Excur- 
sus B. 2 IH, 933-8, &c.; cf. p. 94 above. 

3 'Asim b. 'Adi IS, in/2. 36; WW, 66; cf. IH, 494. 

4 'Abdallah b. Ka'b al-Mazini was 'dmil for maghdnim at Badr IH, 457 ; WW, 
70; IS, iii/2. 73. Budayl b. Warqa' (Khuza'ah) was in charge of prisoners after 
Hunayn WW, 368. 

5 'Abdallah b. Rawahah IH, 177; WW, 286. 

6 'Amr b. Sa'id b. al-'As and Yazid b. Abl Sufyan Caetani, ii. 50 f. (from 
al-Baladhuri, 34), 358, nos. n, 13 (from al-Ya'qubi, ii. 81, where 'Amr is said 
to be over qurd 'arabiyah). 7 WW, 326, 391 ; &c. 


leaders in battle. 1 When it came to the matter of collecting the tax (or 
legal alms) from these tribes, it was to the summoners that Muham- 
mad turned, or at least to the most responsible of them. 2 This 
happened in the case of the tribes of Aslam, Juhaynah, and Ka f b ; 
and the collector for Aslam also collected for the related tribe of 
Ghifar. Certain tribes which had not been in alliance with Muham- 
mad for a long time had collectors sent to them belonging to 
Quraysh or the Ansar; for example, Sulaym, Fazarah, al-Mustaliq. 
The sending of one of the Ansar to Muzaynah is probably different, 
since he was of the Aws, and so an old confederate; the reason may 
have been that no man of Muzaynah was capable of taking the 
responsibility, since even their leading man, Bilal b. al-Harith, 
could be given the comparatively menial task of looking after the 
grazing land for Muhammad's war-horses. 3 The general picture 
is thus one of Muhammad making use of capable men from the 
friendly tribes men who already had a high standing in their 
tribes and in the case of other tribes appointing as his agents men 
of administrative ability from Mecca and Medina. 

The stronger tribes in the centre and north-east of Arabia, in so 
far as they entered into alliance with Muhammad, negotiated 
through their chiefs, or at least through men who aspired to be 
chiefs. To begin with they probably made no contributions to 
Muhammad's exchequer, so that the question of tax collection did 
not arise. Where it did, the chief was responsible. 4 It was therefore 
only in east-central and south-west Arabia that there was scope 
for establishing a system of provincial administration. Muhammad 
had there to deal with a large number of small units, some friendly 
and some unfriendly. He made some use of local men of influence, 
but for the most part he preferred to employ Meccan or Medinan 
'agents'. Some of these were apparently responsible for maintaining 
order and collecting money due to Muhammad, each in a definite 
area. At least one, however, Mu'adh b. Jabal, had duties through- 
out the Yemen and Hadramawt in the regions assigned to other 
agents; these duties included the giving of instruction in the 
religion of Islam and, at least in some areas, the collection of taxes, 
but there is no mention of supervising the local agents. 5 During 

1 Cf. WW, 358. 2 WW, 385; Excursus H. 

3 WW, 184; cf. poverty of 'Amr b. 'Awf al-Muzam, WW, 392. 

4 Cf. cases of Tayyi' and Tamim in Excursus H. 

5 Tab. 1852 f.; IS, i/2. 20. 


the wars of the Riddah these 'agents' commanded armies of which 
at least the nucleus came from Medina. Prior to this, however, they 
normally had no military support beyond that of a dozen or score 
of their fellow citizens of Medina or Mecca. If necessary, they 
could presumably appeal for help to the pro-Medinan party in 
each district. 

Mecca was in a special position. As already noticed, Muhammad 
appointed as his representative there a young man of the clan of 
'Abd Shams, 'Attab d. Asid, but we can only guess at his func- 
tions. 1 There are also references to some subordinate posts in 
Mecca: one man was inspector of markets, another was charged 
with delimiting the sacred area, and a third had some unspecified 
functions. 2 

Thus Muhammad's administrative appointments illustrate the 
nature and extent of his power. In theory he is simply the foremost 
of a number of equal allies. His primacy comes from his office of 
prophet and from the fact that many of the allies undertook to obey 
him. The men whom he sends to perform various functions are 
not officials of an impersonal state, but ' agents' of Muhammad, 
doing what he was formally entitled to do or what his personal 
influence allowed him to do. They probably worked more by per- 
suasion than by coercion. So long as Muhammad lived, his personal 
influence must have seemed to contemporaries to be the cement 
which held the structure together. Yet the building was more 
firmly constructed than appeared and less dependent 'on Muham- 
mad's person ; and later events showed that it was capable of being 
expanded into the administration of an empire. 


The political thinking of the Arabs of Muhammad's time had 
as its centre the conception of the tribe. The tribe was essentially 
a group based on blood-kinship, though in practice this might be 
modified in various ways. Several tribes might take an oath and 
form a confederation; but this was usually only for a limited pur- 
pose, such as fighting against a similar confederation of tribes. 
Again, an individual or a family might for practical purposes 
become a member of a tribe to which he (or it) was not related by 

1 Cf. p. 75 above. 

2 SSL Id b. Sa'id b. al-'A$ (Ibn Hajar al-'Asqalani Al-Isdbah, Calcutta, 1856, 
&c., ii. no. 5083; Usd, ii. 309; cf. IH, 875, killed at af-Ta'if); Tamim b. Asad al- 
Khuza'I (WW, 341); al-Harith b. Nawfal (IS, iv/i. 39). 


blood, as a confederate (halif) or a 'protected neighbour' (jar) or 
a client (mawla). These modifications, however, did not lessen the 
dominance of the conception of the tribe. The tribe was the basis 
of such social security as there was. Only through membership of 
a tribe did life become tolerable for a man, and in return the tribe 
demanded his supreme loyalty. The main tribes were sovereign and 
independent political entities. The Arabs certainly had some super- 
ficial knowledge of the Byzantine, Abyssinian, and Persian empires, 
and they had some idea of kingship and disliked it. 1 This did not 
affect their political thinking, however ; instead they conceived the 
empires in terms of their own tribal system, and, for example, 
made Heraclius, in the story of Muhammad's letter to him, act 
as if he were the chief of an Arab tribe. 2 Thus in studying the 
character of the stata and community created by Muhammad 
which for simplicity will be called the ummah we shall keep in 
mind the question how far the ummah is similar to a tribe in con- 
ception, and how far different. 3 

The outstanding difference is that the ummah was based on 
religion and not on kinship. This idea is nowhere given theoretical 
expression, but it is everywhere implied or assumed. It was implied 
when the Ansar accepted Muhammad as a messenger from God. 
If Muhammad is a messenger, there must be a message; and a 
message in turn implies that God is giving directions to the ummah 
in the practical affairs of life. In many matters of principle Muham- 
mad does not act of his own accord, but merely announces what 
God commands. Thus God is the head and director of the ummah^ 
In the Constitution ( 15) the security enjoyed by members of the 
ummah and groups attached to it is regarded as coming from the 
dhimmah of God, that is, His compact or guarantee of security. 
In the letters and treaties collected by Ibn Sa f d Muhammad's 
name is coupled with that of God. Many groups are given or offered 
the dhimmah of God and of Muhammad ; 4 some are said to have the 
security (aman) of God and Muhammad ; 5 and a few (apparently 
all Christian) have merely ' neighbourly protection' (jiwdr) y but in 
the same two names. 6 This usage continued long in certain spheres ; 

1 Cf. MJ Mecca, !$.; also Ryckmans, Institution Monarchique, for South 
Arabia. * IS, 1/2. 16 ( 2). 

3 Cf. Bertram Thomas, The Arabs, London, 1937, 125, 'super- tribe*. 
* IS, 1/2. 21 ( 16, 17), 23 ( 24, 25), 28 ( 44), 29 ( 45), 34 ( 67), 37 ( 74). 

5 Ibid. 23 ( 2 3 c), 25 ( 30 d), 32 ( 57), 33 ( 61 a, b), 37 ( 75)- 

6 Ibid. 21 ( 14, 15), 29 ( 45), 36 ( iZyjiwdr Allah wa-dhimmat an-nabi) 


the public treasury, for example, was known as 'the wealth of God* 
(mal Allah). 1 

The idea of this theocratic polity is probably not derived directly 
from the Old Testament. It is rather an independent Arabian 
elaboration of certain basic ideas from the Old Testament, such 
as God, revelation, and prophethood. The ummah is not very like 
the Israelite theocracy of the judges. It is closer to the theocracy 
under Moses. The Qur'an, however, shows no detailed knowledge 
of the theocratic government of the Israelites under Moses ; though 
there are many stories about Moses, they tell little about the 
political organization. The ummah, therefore, is not consciously 
based on the community of Moses. The position is rather that the 
Qur'an gives a picture of the relation of prophet and community, 
in which an Old Testament pattern is vaguely to be traced, but 
whose specific colouring is Arabian. We have already noticed one 
Arabian feature of Muhammad's position the parallel between 
the * fifth' which he received and the quarter share usually given 
to tribal chiefs. 

In the Qur'an a development can be observed in the meaning of 
ummah. The word is not from the Arabic root found in umm, 
mother, but is ultimately derived from Sumerian. It appears to 
have come into Arabic at an early period, but whether directly from 
Sumerian or indirectly through Hebrew or Aramaic is not certain. 2 
According to Noldeke's dating of the surahs, most of tfye usages of 
ummah in the Qur'an are in the Meccan period. Richard Bell, on 
the other hand, only regards three instances as certainly Meccan, 3 
though there are a number of others which he describes as 
'Meccan (?)' and 'late Meccan or early Medinan'. It may be there- 
fore that this word was introduced only after the founding of 
a new type of community at Medina was envisaged. Ummah was 
the sort of word that could be given a new shade of meaning; and 
it also was capable of further development subsequently. Hitherto 
it had been said that a prophet was sent to his qawm, but qawm, 
which may be translated 'tribe', had for long been associated with 
the kinship-group, which was the only form of social and political 
organization known in Arabia. When ummah is first used in the 

1 Cf. Q. 24. 33- 

2 Cf. Rudi Paret, art. 'Umma' in El (i); A. Jeffery, Vocabulary, s.v.\ J. Horo- 
vitz, Koranische Untersuchungen, 52. 

3 6. 38 c; 10. 47/48 c; 13. 30/29 D. 


Qur'an it is hardly to be distinguished from qawm\ every beast and 
bird is even said to be an ummah. 1 Mostly, however, the ummah is 
a community to which a prophet is sent; 'each ummah has a 
messenger'. 2 Gradually, however, ummah comes to mean more and 
more a religious community, until in the latest instances (none 
much after Uhud according to Bell) ummah is applied almost 
exclusively to the Muslim, Jewish, and Christian communities, or 
some section of them. Thus the Muslims are told that they have 
been made 'an ummah in the middle', 3 and that they 'have become 
the best community' ; 4 while it is said that among the People of the 
Book is 'an ummah which aims at doing right'. 5 There is a marked 
difference between this later usage and the earlier one according 
to which it could be said that a whole community might reject its 
messenger. 6 % 

What we thus learn from the Qur'an may be supplemented from 
the Constitution of Medina. There in i it is stated that the 
believers and Muslims of Quraysh and Yathrib are one ummah\ 
and this community presumably includes also 'those who follow 
them . . .'. The ummah is thus the complex community at Medina 
to which Muhammad believed himself to be sent. The later article 
( 2 5) which affirms that certain Jews 'are an ummah along with 
the believers', though it could conceivably mean that they consti- 
tuted a community parallel to that of the believers, presumably 
means that they are included in the one ummah. As they are 
specifically allowed to practise their own religion, however, this 
suggests that the ummah is no longer a purely religious community. 
If, however, the last use of ummah in the Qur'an is to be dated 
a little after Uhud while this article is subsequent to the execution 
of Banu Qurayzah (as has been suggested), there is no contradiction 
but only a development dictated by circumstances. 

This development points to a third possible basis for a com- 
munity distinct from kinship and religion, namely, locality. This 
is indeed, if not the whole basis, at least a prominent factor in the 
basis of most settled communities. In the examination of the 
various clans of Medina it was seen that in some cases, like that of 
the 'people of Ratij', organization by kinship appeared to be giving 
place to organization by locality. To the external observer it is clear 

1 6. 38 C. * 10. 47/48 C. 3 2. 143/137 EF. 

4 3. IIO/I06G. 5 5. 66/70 FG. 

6 40. 5 ? c; 28. 18/17 E+ ?; cf. 27. 83/85 E-K 

5783 R 


that the ummah as described in the Constitution of Medina in fact 
has a territorial basis; but it is also clear that this territorial basis 
was not officially recognized by the members of the ummah. The 
ummah has as its core the Muslims now living in Yathrib, but it 
is thought of as a group of clans together with their confederates 
and other * followers'. The basis could never have been purely 
territorial, of course, for nomadic confederates of the Ansar like 
Juhaynah and Muzaynah were presumably included. The terri- 
torial factor appeared again in the caliphate of 'Umar when he 
made all non- Muslims leave Arabia, but, for example, allowed the 
Jews of Wadi '1-Qura to remain since they were not in the Hijaz 
but in Syria. 1 Thus in practice the element or factor of locality 
helps to constitute the ummah, but no recognition is given to this 
in theory where everything is interpreted in terms of the kinship- 

Something of the same kind happens to the conception of hijrah^ 
or 'emigration'. To the European it suggests primarily change of 
location, but the Arab seems to have thought of it rather as a change 
of relationship to one's tribe to make the hijrah was to leave one's 
tribe and attach oneself to the ummah. 2 Two points involving the 
idea of hijrah fall to be considered here. The first is the application 
of the word to members of nomadic tribes who came and settled 
in Medina. There were many of these, and when they pledged 
themselves to Muhammad a distinction seems to have been made 
between the 'nomadic pledge' and the 'pledge of migration' (bay' ah 
'arabiyah, bay' at hijrah). 3 There is no mention of these persons in 
the Constitution, though it is carefully worded and says merely 
'emigrants of Quraysh'; they may have ranked as 'protected 
neighbours' of Muhammad. In the second place the status of 
'Emigrants' or Muhdjirun is given by treaty to the tribes of Aslam, 
Khuza'ah, and Muzaynah. 4 We cannot be certain exactly what this 
involved. For Muzaynah it probably meant that they belonged to 
the core of the ummah in their own right and were not merely 
confederates of the Aws. There is no record of any signal service 
by Muzaynah to justify this reward, but Aslam and Khuza'ah had 

1 Cf. WW, 292. 

* Cf. C. Snouck Hurgronje, Verspreide Geschriften, i. 297-305, 'Twee popu- 
laire Dwalingen verbeterd. I, De hidjra.' 

3 IS, iv/2. 66. 3; cf. p. 86 above; for bay* ah cf. M/ Mecca, 112. 

4 IS, i/2. 24. 17, 25. i4> 38. 13 ( 29, 32 a, 76); p. 86 above. 


served Muhammad well. As explained above, 1 the conferment of this 
status probably also attached these tribes specially to Muhammad 
and so strengthened his position in Medina relatively to the Ansar. 

When we turn from these reflections on the basis of the ummah 
to consider its relations to other groups, we find the conception 
of the tribe very influential. The enemies of the ummah are essen- 
tially the unbelievers and the idolaters, in accordance with the 
religious basis ; but the attitude towards them was that appropriate 
towards hostile tribes. There were few conventions to restrain 
a tribe in its dealings with a hostile tribe, and the individual 
stranger in Arabia in pre-Islamic times had few rights unless 
someone voluntarily gave him protection. But the Muslims dis- 
regarded even the existing rights and conventions where these 
were connected with the old religion, and some of their behaviour 
seemed outrageous to their pagan opponents. 2 Thus the conduct 
of the Muslims to their enemies was that of one tribe to another, 
but with some of the conventions disregarded. Indeed, there was 
nothing in common between the two, no positive relationship, and 
there was no reason for observing any decencies except where non- 
observance debased oneself (like mutilating corpses) 3 or might 
incur unpleasant retaliation from one's enemies. 

The prohibition of intermarriage with pagans is probably an 
indication of this complete separation from idolatry and its adher- 
ents. Soon after al-Hudaybiyah there came a revelation interpreted 
as forbidding Muslim women to remain married to pagan husbands 
and Muslim men to continue to have pagan wives. On being in- 
formed of this command f Umar divorced two pagan women to 
whom he was still married (though they were probably not in 
Medina). 4 The order doubtless had its place in the total strategy. 
Coming after al-Hudaybiyah it emphasized the impossibility of 
being on good terms with Muhammad without accepting his 
claims. It also removed possible sources of false doctrine, and got 
rid of attachments which might have made the prosecution of war 
to the uttermost more difficult. It presumably did not imply the 
prohibition of relations other than marriage with pagan women. 

1 Cf. p. 86 above. The importance of the status is shown by the list of women 
classified as muhdjirdt in IS, viii. 

2 Cf. I. Goldziher, Muhammedanische Studien, Halle, 1888, i. 69. 

3 Cf. IH, 463, 585 J WW, 69, 135 ; &c. 

4 Q. 60. 10; cf. IH, 755 ; WW, 263. IS, iii/i. 190 does not mention the first of 
the two divorced by 'Umar. 


The ummah entered into positive relations with tribes in much 
the same way as a strong tribe did. 'Neighbourly protection* (jiwdr) 
was given in the name of God and of Muhammad. 1 What came, 
however, to be the main conception of the relation of the ummah 
to other groups is that contained in the word dhimmah. This is an 
obscure word in some ways. Its primary meaning seems to be 
'compact* or 'covenant' though this is remote from the meaning 
of the verb dhamma, 'to blame*. It is used twice in the Qur'an in 
the sense of 'compact*. 2 The meaning broadens out, however, to 
that of a compact giving a guarantee of security, and so it comes 
to mean 'guarantee of security* and even 'protection*. Perhaps the 
truth is that our Western minds have failed to seize the essence of 
dhimmah, and so it seems to us that it vacillates between two mean- 
ings. The best we can do is to look at some examples. 

When it says in 15 of the Constitution that 'the dhimmah of 
God is one*, the meaning is presumably that the 'compact guaran- 
teeing security* is one; and this implies (as is shown in the following 
clauses) that all members of the ummah are equally protected, that 
all are equally capable of giving protection which the whole ummah 
is obliged to make effective, and that they all stand to one another 
in the relation of protector and protected, while none is to be 
protected, except temporarily, by anyone outside the community. 
In the letters and treaties of Muhammad collected by Ibn Sa'd 
there is frequent mention of 'the dhimmah of God and,the dhimmah 
of Muhammad*. In many cases the translation 'protection* would 
suit. Sometimes it even seems to be required, as, for example, 
when it is said that 'the Prophet covenants (ahadd) to them (Ghifar) 
the dhimmah of God and the dhimmah of His Messenger*. 3 On the 
other hand, there is sometimes also present the idea of a compact 
or of an obligation binding on God and Muhammad. In the letter 
to the clan Hadas of Lakhm it is said that the Muslim who performs 
his duties is secure 'by the dhimmah of God and the dhimmah of 
Muhammad*, but, if he apostasizes, 'the dhimmah of God and the 
dhimmah of Muhammad, His Messenger* is 'free of (responsibility 
towards) him* (barf ah min-hu)\ and there is the curious addition 
that the man whose isldm is attested by a Muslim is secure 'by the 

1 IS, i/a. 21. 4, 14; 29. 10; 36. 7, 13 ( 14, 15, 45, 72); the last hasjYwar Allah 
wa-dhimmat (Muhammad) an-nabi. Cf. also IH, 986. 10. 

2 9. 8, 10. 

3 IS, i/a. 26. 28 ( 39) = Excvyrsus G, no. 2. 


dhimmah of Muhammad* alone. 1 The reason for the omission of 
God's name here is not clear; there may have been a slight distinc- 
tion present in the mind of Muhammad since in another letter he 
speaks of 'thejiwdr of God and the dhimmah of the prophet'. 2 A 
common expression is that certain people 'have (la-hum) the 
dhimmah\ 3 but in at least one case it is said to be 'upon them* or 
'over them' ('alay-him). 4 The translation 'guarantee of security' 
fits most of these passages, except that, where the dhimmah of 
Muhammad is said to be free of responsibility towards (bari'ah 
miri) someone, it seems to be regarded almost as a part of his per- 
sonal being. 

The explanation of this difficulty in giving a precise meaning to 
dhimmah is perhaps to be found in the fact that the conception 
was a fluid one. One of the chief functions of the tribe or kinship- 
group was to guarantee the security of its members; this was a 
matter of universally recognized custom. Muhammad's problem 
was to find something for his religiously based community that 
would take the place of this customary obligation. He solved his 
problem by developing the pre-Islamic practice of forming an 
alliance or confederacy. The old word halafa, 'formed a con- 
federacy', is used in Muhammad's letter to Nu'aym b. Mas'ud; 5 
but it seems to have been replaced by phrases including aman, 
'security' or dhimmah\ the reason for this change will appear 
presently. The 'pledge of war' at al- f Aqabah was presumably an 
act establishing a confederacy or at least something analogous. The 
confederacy (if we may call it so) thus founded could be joined by 
others, as, for example, Nu'aym b. Mas'ud. Even where no word 
from the root hilf was used, the conditions were those of a con- 
federacy, namely, mutual help and succour. In the early years 
there seem to have been agreements with non-Muslims, presum- 
ably pagans; 6 in the expedition of Hamzah to Sif al-Bahr, perhaps 
the first of all, the tribe of Juhaynah acted as confederates of the 
Muslims; and the letters to the tribes of Damrah and Ghifar 
assuring them of the dhimmah of God and Muhammad, though 
stating their obligation to help Muhammad, do not imply in any 

1 Ibid, 21. 14-19 ( 16) = Excursus G, no. 10; cf. 23. 21 ( 24), the dhimmah 
of Muhammad is bar? ah from whoever disobeys him. 

a Ibid. 36. 13 ( 72). 3 Ibid. 29. 21 ( 46; Excursus G, 5). 

4 Ibid. 27. 6 ( 40); Excursus G, i). 

5 Ibid. 26. 19 ( 35; Excursus G, 8). 

6 Mlthdq in Q. 8. 72/73 seems to be with non-Muslims. 


way that they were Muslims ; the presumption therefore is that they 
were not. Muhammad may have wondered for a time whether 
non-Muslims could have the dhimmah of God, and this may 
explain why in some places we find the dhimmah of Muhammad 
only. But in the end he seems to have extended the dhimmah of 
God to all who belonged to his security system; in later times it 
was precisely the non-Muslim member of an Islamic state who 
was known as a dhimmi or one of the ahl adh-dhimmah (people of 
the dhimmah). Muhammad's hesitation, of course, was only about 
the propriety of God's dhimmah being extended beyond the com- 
munity of Muslims, for he had no hesitation about accepting non- 
Muslims as allies. Moreover, apart from the pagan Meccans at 
Hunayn, there are several instances of men fighting under Muham- 
mad before they became Muslims. 1 ' 

In the old Arab idea of confederacy it was assumed that the 
contracting parties were equal in status, though one of them might 
be stronger than the other. In some of the earlier treaties this 
appears in the form that help (nasr) is due from each to the other. 
In so far as God was mentioned, however, and after Muhammad 
had become the strongest man in Arabia, there was no equality. 
Consequently in the later treaties what is demanded as a condition 
for the granting of the dhimmah of God and Muhammad is the 
fulfilment of the religious duties of Muslims, and in particular 
the payment of zakat\ there are also special demands in certain 
cases. For non-Muslims who make a treaty with Muhammad the 
demand is for what is usually called jizyah. Whatever the name, 
this payment by non-Muslims was regarded as being in return for 
the protection given. It was apparently common in pre-Islamic 
times for weak tribes or settled communities to make payments 
for protection to strong nomadic tribes. Where protection could 
not be given, payment was not accepted; when an individual or 
tribe refused to pay an assessment, protection was withdrawn. 2 
When, in the caliphate of 'Umar, the Muslims had on one occasion 
to retire from Hims they refunded the tax (khardj) which had been 
paid. 3 

1 Cf. p. 89 above; also Abu Ruhm al-Ghifari, WW, 56 f., &c.; Jews, WW, 

2 Cf. D. C. Dennett, jr., Conversion and the Poll-tax in Early Islam, Cambridge, 
U.S.A., 1950, 117- 

3 Al-Baladhuri, 131. 6-9, and Dennett, op. cit. 55-57; the matter is not 
wholly clear, however. 


There does not seem to be any standard term to designate the 
Islamic community after Mecca had been incorporated and many 
tribes had become confederates. Ummah is no longer used in the 
Qur'an or in the treaties. In the latter one occasionally finds terms 
like jama ah 1 or hizb Allah. 2 - An official term was perhaps unneces- 
sary, since diplomacy was carried out in the name of God and 
Muhammad. Presumably members of Muslim tribes in alliance 
with Muhammad were regarded as full members of the community 
(in contrast to non-Muslims). We cannot say at what point this 
change took place and what it involved. After the fixing of stipends 
by 'Umar in 636/15 all full members of the community received 
a stipend; and presumably some such arrangement was in force 
earlier, though nothing is known of details. It would seem that, 
in order to be a ihember of the community, a man must be a 
member of some group which had had a treaty or confederacy with 
Muhammad either of one of the clans mentioned in the Constitu- 
tion of Medina, or of a tribe which had later made a treaty. 
This would explain why there was such eagerness to show 
that every tribe had sent a deputation to Muhammad. It would 
also explain why non-Arabs, on becoming Muslims, had to be- 
come mawdll of Arab tribes. 3 The population of the conquered 
countries as a whole had no treaties with Muhammad or his 
successors, while towns that had treaties had them as non-Mus- 
lims. (We have already noticed how individuals who left their 
tribes and came to settle in Medina were classed as Muhajirun or 

From these points it follows that the Islamic community was 
never thought of as a collection of individuals, but as a collection 
of groups who were in various forms of alliance or confederacy 
with Muhammad. The retention of these groups was perhaps for 
administrative convenience; 4 in the Constitution of Medina the 
clans are responsible for blood-money, ransoms, &c., and later the 
clans and tribes may have been useful, not merely in the adminis- 
tration of justice, but as subdivisions of the stipend roll. Neverthe- 
less there was a strong tendency in Islam to get rid of such 
subdivisions. Every Muslim was to be equally protected by all, 

1 IS, i/2. 27. 9 ( 4i). 2 Ibid. 83. 4 ( 1403); cf. Q. 5. 56/61 ; 58. 22. 

3 Cf. Dennett, op. cit., 58. 

4 Cf. IH, 345 f., where 'Umar inscribes Bilal and all Abyssinian Muslims 
under Khath'am because Bilal was originally 'brothered' with a man of Khath'am. 


said the Constitution, and none was to be protected by others 
from being punished for wrong acts. Thus these subdivisions were 
redundant in theory, and perhaps undesirable in themselves. It 
was a principle attributed probably rightly to Muhammad that 
'there is no confederacy (hilf) in Islam'; that is, no two groups 
within the community were to establish a specially close relation- 
ship. 1 Such a confederacy or special relationship would be tanta- 
mount to a denial of Islam, since it would imply that the protection 
given by Muhammad or his successors was incomplete. (In the 
confederacy of a small group with Muhammad 2 there is of course 
no contradiction; but it was clearly advisable not to use the term 
hilf for the relationship of a group to Muhammad in view of the 
prohibition of hilf in other forms.) 

The institution of the nuqaba* (sing, naqib) 'or 'representatives' 
at the second 'Aqabah is so obscure that it adds little to our know- 
ledge of the nature of the ummah, but it at least confirms that it 
was thought of as a group of clans. The nuqabff were not just 
twelve men representing the Ansar, but were representatives of 
clans; thus As'ad b. Zurarah was the naqlb of an-Najjar, and, when 
he died, soon after the Hijrah, the clan set about appointing a 
successor, but were persuaded to accept Muhammad, who was 
related to them through his paternal great-grandmother. The 
distribution of the 'representatives' among the clans 3 was based 
on the numbers and quality of the members of the claji present at 
al-' Aqabah. No clan had more than two 'representatives'; but, 
while Bayadah with three men present had no naqtb, Sa'idah with 
only two men present had two. The nuqabcf were the leading man 
or men in each clan at al-'Aqabah, and were of course all Muslims. 
A weakness of the institution was that some of the leading men of 
Medina were excluded not merely uncertain supporters of Mu- 
hammad like Ibn Ubayy, but even Sa'd b. Mu'adh, who from 
Badr until his death was the most important of the Ansar. This 
failure to include the real leaders probably explains why we 
hear nothing of the workings of the system of nuqaba? ; it did not 

The practice of what may be called 'brothering' (mu'akhah) does 
not help in the understanding of the nature of the community, but 

1 Cf. Goldziher, op. cit. i. 69; cf. Constitution, 12; also al-Bukhari, Adab 
(?8), 67; Ibn Flanbal, Musnad, i. 190. 2 Cf. p. 245 above. 

3 Cf. table on p. 180; for the institution tfnuqabd' v. M/ Mecca, 145-8. 


it may be mentioned here. The main instance of 'brothering' was 
before Badr, and probably just immediately before it; each Emi- 
grant was paired with one of the Ansar, and the pair who thus 
became brothers were supposed not to leave one another during 
the battle. The purpose of the device was doubtless to prevent the 
different sections of the force from reacting differently to the onset 
of the enemy. If one 'brother' was killed, the other was supposed 
to inherit. Sa'd b. ar-Rabi' (Ba'1-Harith), with whom 'Abd ar- 
Rahman b. f Awf was paired, offered his 'brother' half of his wealth 
and one of his two wives. 1 Perhaps owing to the difficulties about 
inheritance, the practice of 'brothering' was abrogated, but it is 
not clear whether this happened soon after Badr or later. 2 Though 
even Ibn Ishaq 3 speaks as if the 'brothering' before Badr was the 
only instance of the practice, this was not so. Muhammad also 
'brothered' some of his Meccan followers with one another, as we 
learn from the individual biographies in Ibn Sa'd, and this pre- 
sumably happened before the Hijrah; thus 'Abd ar- Rahman b. 
'Awf was 'brother' of Sa'd b. Abi Waqqas as well as of Sa'd b. 
ar-Rabi'. 4 There must also have been some 'brothering' later in the 
Medinan period, since there are cases involving men who were 
not in Medina until after Badr. 5 Further, on two occasions the 
leader of an expedition is said by al-Waqidi to have 'brothered' his 
men to prevent them following the enemy too far; the 'brothers' 
were told not to separate from one another, but in both cases one 
was disobedient and pursued the enemy some distance. 6 These 
latter cases make it probable that the early 'brothering' in Medina 
of Emigrants and Ansar aimed at securing greater cohesion in 
battle. The practice may be regarded as an adaptation of the pre- 
Islamic confederacy to lessen the disadvantages from the con- 
tinuing influence of the kinship-group. 7 

1 IS, iii/i. 89; cf. Q. 4. 33/37, abrogated by 33. 6, according to Tab., Tafstr, v. 
31-35; also C. van Arendonk, art. 'HilP in El (i) 

2 Ibid. 100. 19-25, iv/i. 23. 27; cf. Q. 8. 75/76, 33. 6, and Abu Ja'far an- 
Nahhas, K. an-Ndsikh wa-'l-Mansukh, 159 (on Q. 8. 75/76). 

3 IH, 344-6; cf. phrase win ahl Badr in IS, iii/2. 23. 13, 24. 7. 

4 IS, hi/ 1. 89; but cf. also IH, 934. i, 'brothered' with 'Uthman b. 'Affan. 

5 IH, 933 f., Mu'awiyah and al-Hutat b. Yazid (Tamlm); cf. IS, iv/i. 12. 12, 
al-'Abbas and his nephew, Nawfal b. al-Harith. 

6 WW, 297, 318, in 628/7 and 629/8 respectively. 

7 Cf. J. Schacht, art. 'Mirath' in EI( S) ; also J. Wellhausen, 'Die Ehe bei den 
Arabern' in Nachrichten der kgl. Gesellschaft der Wissenschaften zu Gottingen, 
1893, 461 ; and I. Lichtenstadter in Islamic Culture, xvi (1942), 47-52. 



The finances of the Islamic community are worthy of some 
consideration in detail. During the Meccan period the community 
had nothing resembling public finance, though Abu Bakr and 
perhaps others spent money in freeing slaves who became Mus- 
lims. 1 Several of the articles of the Constitution of Medina deal 
with financial matters, but all except one exemplify the principle 
that each group was responsible for its own expenses and not for 
those of other groups united with it in the ummah; this applies to 
the paying of blood-money and ransoming of captives, 2 and also 
to the expenses incurred in campaigns. 3 The exceptional article 
is 1 1 which states that the believers in general give a debtor help 
towards the payment of ransom or blood-money. How this worked 
in practice is not clear; perhaps it was left to some of the richer 
Muslims to give help, and, if they were unwilling, Muhammad 
himself may have stepped in. When 'Abdallah b. 'Amr (Salimah) 
was killed at Uhud, he left debts equivalent to two years' produce 
of his date-palms ; when the son explained the difficulty to Muham- 
mad, the latter helped him to meet his creditors (though as the 
story stands it does not concern public finance, since Muhammad's 
method was the miraculous multiplication of the stock of dates). 4 
Even if money had been given in this case, it would still show 
a rudimentary financial organization. 

It is not clear how any of the Emigrants made living in the 
period between the Hijrah and the battle of Badr, far less con- 
tributed to a common purse. They presumably did not take up 
agriculture, and they presumably did not plan to live indefinitely 
off the hospitality of the Ansar. It has been suggested that they 
must have had in view either raids on Meccan caravans or long- 
distance trading, and both courses would lead to conflict with 
Mecca. 5 There is some evidence that the Emigrants engaged in 
commercial operations. ' Abd ar-Rahman b. 'Awf rejected the offer 
of half the wealth of his Medinan 'brother', and asked instead to 
be shown the market (sc. that among B. Qaynuqa'); he soon 
returned with a skin of butter and a cheese that he had gained by 
his superior business acumen. 6 This presumably happened soon 

1 IS, iii/i. 122; cf. M/Mecca, 118. 2 2-10. 

3 37 > 2 4( = 38) is doubtless to be interpreted in the light of 37. 

4 IS, iii/2. 107. 5 M/Mecca, 148. 
6 IS, iii/i. 88; al-Bukhri, Buyu' (34), i. 

vii. 4 FINANCE 251 

after the Hijrah and certainly before the expulsion of B. Qaynuqa* ; 
the next two pieces of evidence to be mentioned cannot be dated 
certainly, but both of them are probably to be referred to this early 
period. 'Umar did not hear an instruction given by Muhammad 
because he was engaged in a market transaction. 1 'All obtained 
rushes from one of the Ansar in order to sell them. 2 'Uthman and 
'Abd ar- Rahman b. 'Awf traded in the captives of B. Qurayzah. 3 
In general the Emigrants are said to have spent their time in the 
markets while the Ansar were in their fields, 4 though commerce 
was not unknown among the Ansar. 5 At a later period the Muslims, 
presumably both Emigrants and Ansar, did some trading on 
the expedition of Badr al-Maw'Id in 626/4 ; 6 Zayd b. Harithah 
attempted to take a trading caravan to Syria in 627/6, but was 
ambushed; 7 Dihyah b. Khalifah al-Kalbl had merchandise with 
him when Muhammad sent him to 'Caesar'. 8 

From all this it may be concluded that the Emigrants gained 
what they could by trading, both before Badr and after. It must 
not be supposed that every Emigrant could earn his supper as 
easily as 'Abd ar- Rahman; he was the merchant par excellence 
among the Muslims, and became extremely wealthy under the 
caliphs. .With the clan of Qaynuqa' already, at least in part, de- 
voted to trade, there cannot have been a livelihood for seventy 
Muslims in the early months, though things would be easier after 
the expulsion of Qaynuqa'. At least until Badr, then, the Emigrants 
must have been partly supported by the Muslims of Medina. 
Indeed, one of the complaints made to Ka'b b. al-Ashraf by the 
men who assassinated him, to persuade him that they were not 
Muslims, was that Muhammad and his Meccan followers were 
burdensome to the people of Medina. 9 Presumably Muhammad 
made no specific demands on the Ansar, but merely exhorted them 
to 'contribute* what they could. The process seems to be described 
in a verse of the Qur'an: 10 

The good which ye contribute is ... for the poor, who have been 
restricted ( ? fighting) in the way of God, and are therefore unable to 
knock about in the land (to trade) ; the ignorant think them rich because 
of their self-restraint; but one may recognise them by their mark; they 
ask not importunately of the people. 

1 Al-Bukhari, ibid. 9. 2 Ibid. Sharb (42), 13. 3. 

3 WW, 221. 4 Al-Bukhari, Buyu' (34), i. 

5 E.g. ibid. 9. 6 WW, 168. 7 Ibid. 238. 

8 IH, 976. 2. 9 Cf. p. 210 above. I0 2. 273/274 E. 


The Qur'an has ample evidence of the importance of voluntary 
'contributions' in the plans for the young community at Medina. 
Men are commanded to believe in God and His messenger and 
contribute of their wealth. 1 Their contributions are a loan they 
lend to God; He knows what they do; He will repay them the 
double and more. 2 They are to contribute what they can spare. 3 
From the very first the Qur'an had insisted on generosity; 4 but it 
is apparently only in the early Medinan period that the requests 
for contributions commence. Though they were voluntary, and 
though men were encouraged to give them by the promise of a 
reward from God, yet there must have been some pressure or 
obligation on the richer Muslims to contribute, since the with- 
holding of contributions (bukhl, &c.) is criticized and threatened 
with Divine punishment. 5 There is also mention of some persons, 
presumably Hypocrites or Jews, who made contributions but in 
other ways did not act as true believers. 6 In the difficult period 
after Uhud some of the opponents tried to stop the contributions 
(sadaqdf) by mockery. 7 

One or two incidents are recorded which show the kind of thing 
which happened. On the expedition to Sif al-Bahr in November 
629 (vii/8) when food was scarce, Qays, the son of Sa'd b. 'Ubadah, 
on three successive days bought a camel (by promising to pay in 
dates on his return to Medina) and had it slaughtered for the party. 8 
About a year later Muhammad asked for gifts towards the fitting 
out of the great expedition to Tabuk with arms, camels, and pro- 
visions, and the leading Muslims responded generously. 9 A note 
has been preserved of large sums which ' Abd ar-Rahman b. 'Awf 
contributed (tasaddaqd) during Muhammad's lifetime. 10 A man of 
the clan of an-Najjar presented (tasaddaqd) a stronghold (qasr) to 
Muhammad. 11 These are no doubt some of the* high-lights' among 
the contributions, but they show how the early Muslim com- 
munity lived. Those who had anything to spare were expected to 
give to those in need, but there was no regimentation. Gradually 

1 2. 195/191?; 2. 254/255 E; 57. 7 E. 

2 2. 245/246 E; 5. 12/15 F; 57. ii, 19/18 E-; 64. 170; 73. 20 E-f ; 2. 261/263 
? FG; &c. 3 2. 2 1 9/2 1 6 f. ?FG. 4 Cf. M/ Mecca, 68 ff. 

5 3. 180/175 G; 4. 37/41 ff. G, rejecting the interpretation in IH, 389 f. in 
favour of that of some commentators in Tab., Tafsir, ad loc. (v. 51). 

6 2. 262/264 ff. ? FG; cf. 9. 53 f. i. 

7 9. 75/76 ff. GH; cf. 4. 37/41 G. 8 WW, 317 f- 

9 Ibid. .101. I0 Usd. iii. 116. i ff. " Tab. i. 1528. 4. 

vii. 4 FINANCE 253 

the force of public opinion must have compelled the meaner among 
the men of affluence to make some contributions. Where necessary, 
Muhammad may have dropped a hint, as he is said to have done 
on one occasion to 'Abd ar-Rahman b. 'Awf, 1 and as he did before 
the expedition to Tabuk. 

What has been said so far is based mainly on those Qur'anic 
verses which contain the word nafaqah or phrases such as ma 
anfaqtum, these must be translated 'contribution', 'what you con- 
tribute', &c., although the common meaning is 'expenditure* and 
'what you spend'. Some verses with sadaqah (alms) have also been 
used, since there are passages where both roots occur without any 
apparent difference of meaning. In the early Medinan period the 
word zakdt also seems to refer to contributions or voluntary alms- 
giving; at least therfe is no evidence to show that at this period the 
zakdt, which later became the 'legal alms', was a fixed proportion 
of a man's property or income. An examination of the use of this 
word in the Qur'an 2 reveals that it was in some way closely con- 
nected with the Jews, though not used exclusively with respect to 
them. One inference from this is that the Jews of Medina were 
also expected to make contributions. For the most part these were 
presumably to take the form of alms to the poorer members of the 
community, among whom Emigrants would be prominent. Shortly 
before their expulsion an-Nadir were asked to contribute to a pay- 
ment of blood-money. 3 

The problem of how zakdt came to have its technical meaning 
of 'legal alms' lies outside the scope of a life of Muhammad, since 
the change came about later. What must be considered, however, 
is the extent to which there was a transition from voluntary contri- 
butions to fixed 'alms' during Muhammad's lifetime. Definite pro- 
portions of property or income to be paid as 'alms' to Muhammad 
or one of his agents are mentioned in some of the treaties from the 
last two years or so of his life. The usual word, if there is any at all, 
is sadaqah, and the collector of these amounts is consequently 
known as musaddiq. This, according to available evidence, is the 
beginning of the system of 'legal alms', for which zakdt became 
the usual name, while sadaqah came to be reserved for voluntary 
or supererogatory almsgiving. There are no grounds, however, for 
thinking that the fixed proportions were made obligatory for all 

1 IS, iii/i. 93. 2 Cf. Excursus I. 

3 IH, 652; WW, 161; cf. p. 211 above. 


Muslims during Muhammad's lifetime. Not even all the nomadic 
tribes in alliance with him are said to have given sadaqdt. On the 
contrary, the presumption is that completely voluntary almsgiving 
remained the rule for those Muslims for whom no fixed sadaqah 
was prescribed by agreement. This presumption is supported by 
the appeal for contributions before the expedition to Tabuk. After 
the battle of Hunayn the major part of the income of most of the 
Emigrants and Ansar probably came from sharing in the spoils 
and booty of the various expeditions, and was not therefore liable 
to 'legal alms'. (The same consideration would apply even more 
widely during the caliphates of Abu Bakr and c Umar.) Thus, in 
a sense, there is no transition from voluntary contributions to 
legal alms. The older Muslims continued to act as they had been 
doing; but fixed sums were prescribed for the later additions to 
Muhammad's security system. 

The question of gifts is relevant to this matter of voluntary or 
stipulated alms. As early as the expedition of al-Hudaybiyah 
Muhammad is said to have refused to accept milk as a gift from 
some pagans, but to have paid for it. 1 At al-Ji'ranah a man of Aslam 
presented some sheep to Muhammad, who made to refuse them, 
until the man assured him that he was a Muslim and had paid his 
tax to the collector for his tribe. 2 The latter point is significant. 
It suggests that Muhammad was not prepared to accept voluntary 
alms in cases where a man might claim that this exempted him 
from the fixed 'legal alms', which was doubtless heavier. On 
another occasion, presumably after Hunayn (though placed earlier 
by al-Waqidi because of the connexion with 'Uyaynah b. Hisn), 
Muhammad stated publicly that he would accept gifts only from 
Quraysh and Ansar. 3 This further suggests that only these two 
groups were permitted to continue giving voluntary contributions. 

It is also possible that there was little difference between the 
obligatory payments made to Muhammad by Muslims and those 
made by Christian Arabs. At least in some cases, however, a special 
name was given to the latter, jizyah, derived from a Qur'anic verse 
(9. 29), which bids the Muslims fight against the Christians until 
they are subdued and pay thejizyah. The word is thought to have 
come from a similar Syriac word meaning 'poll-tax', either directly 
or through Persian, but it might also be an Arabic formation mean- 
ing 'due' or 'satisfaction', and something of this was doubtless 
1 WW, 242. 2 Ibid. 374- 3 Ibid. 232. 

vii. 4 FINANCE 255 

suggested when it was used in the Qur'an. 1 Despite the different 
name, however, the same men often collected both sadaqdt and 
jizyah. 2 It is also possible that in the early days some Christian 
groups had to pay a tax levied on the herds or fields like the 'legal 
alms', and not on the heads like the jizyah. 3 

Other names are also occasionally used for the fixed sums of 
money or goods which had to be paid to Muhammad. In one 
passage a man is empowered to collect 'ushur, 'tithes', as well as 
sadaqah\ this may have been a pre-existing tax that was continued 
under Muslim administration, but it is impossible to be certain. 4 
Another word found is si' ay ah. 5 Its precise significance is * money 
earned by a slave which counts towards buying his freedom*. It 
occurs in connexion with some of the tribes defeated at Hunayn, 
and probably implies that they were required to pay a fixed sum for 
a number of years by way of ransom, but not in perpetuity. 

In voluntary almsgiving the gift presumably went direct to the 
recipient, since secret almsgiving is regarded as possible. The fixed 
sadaqah, however, and other taxes and dues were paid to Muham- 
mad, and thus belong to public finance. From an early period 
Muhammad had been holding public money or goods, namely, 
the khums or 'fifth' of the spoils or booty captured on campaign. 
The standard account is that the 'fifth' was not instituted until 
after Badr, and was first applied in the campaign against B. Qay- 
nuqa'. 6 At Badr there was a dispute about the division of the spoil, 
and a verse (8. i) is said to have been revealed giving Muhammad 
power to dispose of the whole as he pleased. After this it seems to 
have been agreed that a fifth of the total should go to Muhammad 
for public purposes, and that the remainder should be divided 
equally among the participants; a horseman received two shares 
for his horse in addition to his own share. The 'fifth' was parallel 
to the fourth part that it had been customary for the chief of a tribe 
to receive. 7 Again like the chief of a tribe Muhammad was entitled 
to have a 'first pick' (soft) before the general distribution; in pre- 
Islamic times this usually was some object such as a she-camel, 
horse, sword, or girl, and Muhammad is said to have chosen 
a sword at Badr. 8 Muhammad also received a share of the booty 

1 Jeffery, Vocabulary, s.v.; Lane, s.v.; cf. IS, i/2. 28. 2, jizyat ar$-hd. 

2 IS, 1/2. 20, 28, &c. 3 Excursus G. 

4 WW, 65, 93; cf. Bell on Q. 59. 7 F; he connects w. 6, 8-10 with an-Natfir; 
cf. also Dennett, op. cit. 21. s Cf. p. 101. 6 Cf. p. 232. 


(presumably three if he was on horseback) along with the others. In 
his negotiations with nomadic tribes Muhammad regularly insisted 
that they should pay the * fifth', and his share (sahni) and the safi 
are often mentioned as well. 1 

The actual division of the spoil was efficiently carried out, as was 
to be expected in a capable commercial community. When the 
booty consisted of sheep and camels, it was a simple matter to give 
so many to each man, reckoning ten sheep equal to one camel. 2 
When the booty was more varied, as after Hunayn, it was divided 
into lots regarded as equal in value. Each man received a particular 
object or objects, but there were usually dealers about, and he had 
no difficulty in exchanging his goods for cash; 3 at Khaybar the 
booty was auctioned to the campaigners and (presumably) dealers. 4 
Thus the commercial transactions resulting from the division of 
booty must latterly have employed many men, apart from those 
entrusted by Muhammad with guarding and dividing it. 
Muhammad had a special agent for the 'fifth', Mahmiyah b. Jaz' 
az-Zubaydi, a brother-in-law of Muhammad's uncle al-'Abbas. 5 

When the Muslims gained booty not in actual fighting but 
through an agreement, Muhammad claimed the whole. This hap- 
pened in the case of the Medinan Jewish clan of an- Nadir. By 
arrangement with the heads of the Ansar the lands of an-Nadir 
were not divided equally among all the Muslims, but were given 
to the Emigrants (along with two of the poorest of the Ansar); 
the Ansar, however, insisted on the Emigrants continuing to live 
with them, perhaps to avoid dispersing their forces and thereby 
weakening them. 6 There is no mention earlier of any use of the 
habitations of Qaynuqa'; perhaps Arabs shared the strongholds 
with them. There is also no mention of how the lands of an-Nadir 
were cultivated by the Emigrants. Some sources imply that at 
Khaybar and Fadak also the lands fell entirely to Muhammad. 7 
Whatever the precise legal position, a new system was instituted 
at Khaybar, which was followed in the conquests after Muham- 
mad's death. The previous owners were allowed to remain in 
occupation but had to hand over half of the produce to the Muslims. 

1 IS, i/2. 30. 7 ( 48); contrast ibid. 25. i, 2 ( 30 d). 

2 WW, 226, 387, &c. 

3 Ibid. 282, 284. 

4 Ibid. 275, 281 ; there are apparent contradictions in the accounts of Khaybar. 

5 IH, 783; WW, 177, 221, &c.; IS, iv/i. 146. 5. 

6 WW, z66. 7 Ibid., but cf. 286, mention of 'fifth*. 

vn. 4 FINANCE 257 

The assessment and collection of the half was the work of an over- 
seer, presumably appointed by Muhammad. From the 'fifth', or 
perhaps rather third, of the lands retained by Muhammad he 
assigned so many loads of dates and grain annually to his wives 
and to members of the clans of Hashim, al-Muttalib, &c. The 
quantities varied from five to two hundred loads. The remainder 
of the lands of Khaybar was divided into eighteen lots and assigned 
to Emigrants and Ansar. The lands of Wadi '1-Qura were also 
divided. 1 The capture of Khaybar thus made a vast difference to 
Muhammad's financial position; his responsibilities, of course, 
were also growing. (An interesting measure of Muhammad's grow- 
ing wealth is the number of horses on his expeditions ; at Badr in 
624/2 there were over 300 men and only 2 horses; at Badr al- 
Maw'id in 626/4 there were 1,500 men and 10 horses; at Khaybar 
in 628/7 there was about the same number of men, but 200 horses; 
at Hunayn in 630/8 700 Emigrants alone had 300 horses, and 4,000; 
Ansar had another 500 ; finally on the great expedition of Tabuk in 
630/9 there are said to have been 30,000 men and 10,000 horses. 2 
The military importance of these figures can be seen from the fact 
that at Uhud the Meccan cavalry, which played a decisive part in 
the battle, numbered only 200 in a force of 2,000. ) 3 

Other miscellaneous sources of revenue may be briefly listed. 
A Jew called Mukhayriq fought with the Muslims at Uhud 
and fell, having previously willed his property to Muhammad. 4 
Muhammad seems to have had sole disposal of the temple 
treasure from at-Ta'if ; this was perhaps specified in the treaty 
when the inhabitants consented to the destruction of the goddess. 5 
Finally, after the conquest of Mecca Muhammad demanded 
* loans' from some of the rich Meccans; part of the money was 
used to pay the damages due to B. Jadhlmah. 6 As there had 
been no plundering in Mecca, this was no doubt considered a 
fair arrangement. 

These, then, so far as our records go, are the various sources of 
the income which came to Muhammad to be spent for the public 
weal. 7 The spending does not require much discussion. 

The Qur'an contains several sets of directions for the spending 

1 IH, 773-6; WW, 285-96. * WW, 39, '68, 285, 358, 395- 

3 Ibid. 102. 4 IH, 354; WW, 124. 

* WW, 384 6 Ibid. 348, 353- 

7 For the name mdl Allah, 'wealth of God*, applied to the public treasury, 
cf. H. Lammens, in Melanges de la Facultt Orientate de Beyrouth^ vi (1913), 403 f. 

5783 S 


of money, which resemble one another although they apply to 
different things. Some are instructions to individuals about the use 
of their private wealth ; they are to bestow it on relatives, on orphans 
(who are probably also relatives), on the poor, on the 'son of the 
way 1 , and, in one passage, on beggars and for the ransoming of 
captives. 1 One of these passages deals explicitly with contributions, 
but it is not concerned exclusively with the needs of the Emigrants, 
for it mentions parents as recipients ; thus the system of voluntary 
contributions would seem to have grown out of general principles 
for the use of wealth. What is curious, however, is that when 
regulations for the use of the 'fifth* appear in the Qur'an, the 
groups mentioned are the same relatives, orphans, the poor, and 
the 'son of the way'. 2 Finally, when at a comparatively late period, 
regulations are given for the use of the sadaqat fixed by treaty, the 
poor (and destitute) and the 'son of the way* are again named, 
though they are now accompanied by the agents for the sadaqat, 
'those whose hearts are reconciled', slaves (for their liberation), 
debtors, and expenditure 'in the way of God'. 3 

These instructions and regulations are an impressive witness to 
the continuity of the Qur'anic message. From the very first it had 
been implied that at the root of many of the social evils of the day 
was a false attitude to wealth. 4 Generosity had been urged on the 
believers, and niggardliness had been denounced as leading to Hell. 
By reading between the lines it can be seen that the evils of the time 
are linked with the growth of individualism. Men think of them- 
selves primarily as individuals and not as members of a tribe or 
clan. Consequently they become selfish and neglect their traditional 
obligations to fellow tribesmen and clansmen, even to members of 
their own family. The classes on whom a man is to bestow his 
wealth are presumably those that were the chief sufferers from the 
breakdown of the tribal system. Orphans are prominent because 
men frequently died young and left small children, and it was easy 
for the guardian to appropriate any property the father had left. 
Relatives, other than orphans, are indigent members of the same 
clan, who under the old system would have been cared for by 
the chief or some wealthy clansman, but were now left to fend for 
themselves. The poor were probably at first those inhabitants of 
Mecca or Medina who had no clear connexion with any clan, or 

1 2. 177/172 F; 2. 215/211 FG; 4. 36/40 E ; 17. 26/28 CE+. 

* 8. 41/42 F+. 3 9. 60 i. 4 MIMecca 68-71, 72 ff. 

VH. 4 FINANCE 259 

whose clan was too poor itself to do much for them; this class 
would thus only come into being in so far as groups based on 
locality replaced kinship-groups. The mention of the (two) parents 
may be due to cases where strong men neglected their parents 
when they became old and weak; but it is more likely that its aim 
was to correct some anomaly of the obsolescent family and kinship 
system (to be considered in the next chapter). 

The 'son of the way' is a problem. It is natural to interpret the 
phrase to mean 'traveller', but it is difficult to see why there should 
be so much insistence on helping travellers. Inhospitality can 
hardly have been a great social evil in Medina, where all or most 
of the passages with the phrase were revealed. Another suggestion 
is that it means 'guest' (dayf), and that, when a guest stayed more 
than the three days during which it was obligatory to look after 
him, his entertainment became a sadaqah. 1 On this interpretation 
the hospitality given to Emigrants by the Ansar could be counted 
as fulfilling the religious obligation to be generous with one's 
wealth. It is unlikely that any more precise group should have 
been meant and that all memory of this should have been lost. 2 

The latest regulations those for the sadaqdt show how Mu- 
hammad's responsibilities had grown. He now requires agents to 
collect the sadaqdt from his 'allies', and these agents have to be 
paid. Affairs of state also require the spending of money, and, since 
the state is a theocracy, this may simply be described as 'expendi- 
ture in the way of God'. The primary reference is doubtless to 
military expenditure, since 'fighting in the way of God' was a 
common phrase. 'Those whose hearts are reconciled' seem origin- 
ally to have been the members of the deputations which came to 
Medina in a stream after the battle of Kunayn ; Muhammad was 
in the habit of giving each man a present of several ounces of 
silver. 3 

Scattered in the sources are various illustrations of Muhammad's 
practice. He had some of the booty and captives from B. Qurayzah 
sold in Syria and the proceeds used to buy arms and horses ; 4 and, 
as already noticed, contributions were requested to equip the 
expedition to Tabuk. Again, besides his gifts to members of 
deputations, to women who accompanied expeditions and others 

1 Tab., Tafsir, ii. 55, on Q. 2. 177/172. 

2 But cf. Bell, Translation, i. 24, n. 3. 

3 Cf. below, p. 349, and IS, i/a passim. 4 IH, 693; WW, 221. 


all probably regarded as obligatory under Arabian conditions we 
find him giving large subsidies to certain tribes. B. Ju'ayl (of Bali) 
received the si'ayah or liberation payment Muhammad had im- 
posed on tribes defeated at Hunayn, and the tribute from the Jews 
of Maqna was assigned to men of Sa r d Allah and Judham. The 
aim must have been to strengthen these tribes and to ensure their 
allegiance to Islam. 1 In respect of blood- money, too, Muhammad 
seems to have been responsible where Muslims on an expedition 
killed or injured someone in alliance with him; 2 this is not 
necessarily contrary to what is said in the Constitution about 
blood-money as a clan responsibility. He may also have paid 
blood-money or debts himself, where to leave them to those 
responsible might have occasioned quarrels. 

The provision for his wives and relatives which Muhammad 
made out of the annual tribute from Khaybar might seem to 
indicate partiality. It must be remembered, however, that Muham- 
mad stood in a special relation to the clans of Hashim and al- 
Muttalib; he was their leader among the Muslims. According to 
the principles of the Qur'an, therefore, it was above all to him 
that the poor and needy members of these clans must look for 
help. Moreover, while the leaders of other clans of the Emigrants, 
such as f Abd ar-Rahman b. 'Awf, might spend their time in the 
market making money, some of which they would give to their 
relatives, Muhammad had to devote all his time to political duties, 
and it was thus no more than fair that he should use on their behalf 
some of the money that came to him. As we have seen, the kinship- 
group had still a part to play in the Islamic community, and, where 
a man had wealth to bestow, his relatives had a strong claim on it. 
It would thus not be too much to assert that Muhammad's treat- 
ment of this wives and relatives at Khaybar is intended to exemplify 
the Qur'anic ideal of generosity towards one's kith and kin. 

1 IS, i/2. 24; WW, 405; cf. p. 117 above. 

2 WW, 160, for B. 'Amir; ibid. 353, B. Jadhlmah. 



IT was suggested in Muhammad at Mecca that the proclamation 
of a new religion was a response to the malaise of the times, and 
that this malaise was at bottom due to the transition from a 
nomadic to a settled economy. From the point of view of social 
structure there was a tendency to replace tribal solidarity by 
individualism. Individualism fostered selfishness, and selfishness 
knew very well how to twist nomadic ideals and practices to the 
private advantage of those who found themselves with a measure 
of power. There was a corresponding growth of discontent among 
those who found themselves at a disadvantage in the struggle for 
wealth and power. 

The purpose of the present chapter is to describe the social 
reforms instituted by Muhammad. 1 It is not enough, however, 
simply to describe them in isolation. An attempt must be made 
to see them as an adequate response to the needs of the times. 
Moreover, they will be found to be no sheer novelty but an adapta- 
tion of existing ideas and practices. The relation of the new social 
institutions to the background must therefore be given special 


Prior to Muhammad security of life, in so far as it was main- 
tained, was maintained by the principle of the blood-feud and the 
lex talionis. It must be insisted that this is not a barbarous practice 
to be abolished as quickly as possible, but a form of justice, or at 
least of the prevention of indiscriminate killing. From the stand- 
point of modern society it seems barbarous, but that is because 
it belongs to a level of social organization we have outgrown. 

1 Cf. Robert Roberts, The Social Laws of the Qordn, London, 1925, where 
most of the relevant Qur'anic passages are collected ; they are discussed in the 
light of later practice. (The author published an earlier form of this work in 
German in 1908.) 


Moreover the process of outgrowing it does not consist in uproot- 
ing it, but in transforming it. 

The blood-feud belongs essentially to a society consisting of 
groups, normally groups of kinsmen. When a member of one 
group is killed or injured by a member of another group, the first 
group is in theory entitled to exact an eye for an eye, a tooth for 
a tooth, and a life for a life. The duty of exacting this vengeance 
rests specially on the next-of-kin (of mature age), but he has the 
support of his clan or tribe. Though it is preferable to inflict the 
penalty on the person responsible for the death or injury, it may 
be inflicted on any member of his clan or tribe instead of him. It 
is thus clear that ultimately the responsibility both for the original 
act and for exacting vengeance is Communal. 

These points are illustrated by many stories 6f pre-Islamic days, 
and also by the behaviour of the pagan Meccans and other oppo- 
nents of the Muslim community at Medina. As an example there 
may be cited the gruesome account of what happened to two 
Muslims captured at ar-Rajf in 625/4. Their nomadic captors took 
them to Mecca, and they were readily purchased by the families 
of men killed at Badr. They remained in captivity until the end of 
the sacred month, and were then killed in cold blood. Against one, 
Khubayb b. 'Adi, 1 the first blow was struck by a son of al-Harith 
b. c Amr (Jumah), but as he was a mere child he was unable to 
inflict a mortal wound. 2 Though Khubayb and his companion were 
probably present at Badr, neither seems to have killed anyone there. 
Thus they were paying for the deeds of the Muslims. 

When a member of the offending group had been killed, venge- 
ance was satisfied and the two groups were supposed to live in 
peace. There was, of course, no binding law about this, and, even 
if there had been, no authority to enforce such a law. It was merely 
a generally recognized custom. There was no punishment for not 
observing it, only the disadvantage (if it was a disadvantage) of 
continuing in a state of war with another tribe. In practice the two 
parties frequently disputed about what constituted a fair requital. 
In the well-known story of the war of Basus, the trouble began 
when the chief Kulayb perhaps the most powerful man in central 
Arabia in his day by his pride provoked his brother-in-law Jassas 
to kill him. Kulayb's brother one day ran into a well-born 

1 Apparently sometimes confused with Khubayb b. Asf (or Yasaf); see IH, 
Index, also Usd, and Ibn tlajar, I$dbah. 2 IH, 640 ff.; WW, 158 ff. 


youth of Jassas's tribe and killed him. An influential kinsman of 
Jassas, who had hitherto stood aside from the quarrel, made it 
known that, if the youth's life was accepted as an equivalent for 
that of Kulayb, he would count the matter settled. The insolent 
answer was given, however, that the youth was equivalent only 
to Kulayb's shoe-latchet. Things then went from bad to worse, 
and much blood was shed on both sides over a period of many 

Such developments are perhaps inevitable where there is no 
supreme authority but each tribe is a sovereign political body. 
The best hope of peace lay in the good offices of men friendly to 
both parties. If both could accept someone as arbiter, there was 
some hope of peace. This is how the war of Basus was brought to 
an end. Usually the*number of dead on each side was reckoned up, 
and the tribe with the surplus of losses received from the other 
tribe so many camels for every man of this surplus. The older view, 
however, was that it was dishonourable and a sign of weakness 
to accept camels in lieu of blood, and it was not extinct in Muham- 
mad's time. One of the Ansar, Hisham b. Subabah, was accident- 
ally killed by another. His brother Miqyas came to Muhammad 
and asked for blood-wit (diyah\ which was duly paid. Evidently 
this did not satisfy his sense of honour, however, for, when there 
was an opportunity, he killed the man responsible for his brother's 
death, and fled to Mecca. 1 Nevertheless, the wiser and more pro- 
gressive men of the time seem to have recognized the advantages 
of substituting a blood-wit for the actual taking of a life. An 
unreliable story about 'Abd al-Muttalib, Muhammad's grand- 
father, claims that an action of his in redeeming his son for a 
hundred camels led to the general recognition of a hundred camels 
instead of ten as a proper blood-wit for a man. 2 This may be taken 
as evidence, even if the figures are not accurate, of a tendency to 
raise the blood-wit in order to make the acceptance of it the more 
attractive course. Just before the battle of Badr some Meccan 
opponents of Abu Jahl urged c Utbah b. Rabi'ah of 'Abd Shams 
to declare that he would be responsible for the blood-wit of his 
confederate 'Amr b. al-Hadrami who had been killed at Nakhlah; 
in this way a battle might have been avoided, for according to the 

1 IH, 728, 819; WW, 176. For the idea of the blood-wit as dishonourable, cf. 
R. A. Nicholson, A Literary History of the Arabs, Cambridge, 1930, 93. 

2 IS, i/i. 54; cf. T. H. Weir, art. 'Diya' in El (i). 


traditional view this blood was the only valid cause of quarrel 
between the Meccans and the Muslims. 1 

Such, then, was the system of social security in Arabia as it 
existed at the time of the Hijrah. In Mecca it worked well, for 
Quraysh were noted for the quality of hilm, a combination of 
wisdom, concern for material prosperity, and self-control. In 
Medina it worked badly, and indeed had broken down. In two 
respects the basis of the system was communal and not individual- 
istic. Firstly, the system rested on the principle of communal 
responsibility for crimes. Where there is a strong sense of 
group loyalty it is always possible for the tribe to hide an indi- 
vidual offender, especially in the desert with only camel transport 
available, but it is not easy for a whole tribe to remain in hiding 
indefinitely. Under such circumstances the only way to maintain 
order is to hold the group responsible for the misdoings of members 
of it, and this is in fact done by modern civilized governments in 
circumstances of this kind. Secondly, the system apparently aims 
at maintaining or restoring the relative strength of the tribes in- 
volved in a quarrel. If a member of one tribe is killed, then the 
tribe responsible for his death must be weakened to the same 
extent. This idea recurs in some of Muhammad's regulations. 

The system as a whole had deep roots in Arabian society, and 
there could be no question of replacing it by anything else or even 
of radically altering it. Muhammad may be said to have accepted 
in general the principles underlying the system, and to have set 
himself to reform the most serious abuses. 

In a certain sense Muhammad's greatest innovation was not an 
innovation at all. This was the establishment of a new type of 
group, the Islamic community or ummah, which was based not on 
blood-relationship, but on a common religious allegiance. As has 
been noted in the previous chapter, the ummah was conceived as 
a kind of tribe. The incident just quoted of the killing of Muslim 
captives in Mecca as retaliation for men killed at Badr suggests 
that for purposes such as this the ummah was regarded as a tribe 
even by its enemies. This inference is not certain, however, since 
the two men for whom vengeance was taken seem to have been 
killed by members of the Ansar, and the two captives also belonged 
to the Ansar; consequently the Ansar and not the whole ummah 
may have been regarded as the group responsible. The case of the 

' WW, 50 f. 


two men of B. 'Amir killed by a Muslim fugitive from Bi'r Ma'unah, 
for whom Muhammad felt obliged to pay blood-money, shows that 
he accepted responsibility for the acts of members of the ummah> 
at least towards tribes with whom he was in alliance. 1 In exacting 
vengeance the ummah certainly functioned as a tribe. According 
to the Constitution ( 19) 'the believers exact vengeance for one 
another where a man gives his blood in the way of God*. The 
Qur'anic verse (2. 177/173 E ) prescribing retaliation (qisds) for 
the slain doubtless applied to non-Muslim groups, even if, sup- 
posing the second half not to be a later addition, it also applied 
from the first to killing within the ummah. The following assertion 
that 'in retaliation is life for you, O ye of insight; mayhap ye will 
show piety', is a reminder that the taking of vengeance was a duty, 
and the performance of this duty the mark of a virtuous man. 

It was probably only to tribes with whom he had some alliance 
or agreement that Muhammad would have considered paying 
blood-money. With regard to other tribes it is not clear whether 
he regarded the ummah as bound by the rule that held in internal 
cases, namely, not to exact more than the equivalent by way of 
vengeance. The verse just quoted suggests that the rule also held 
in external cases, for it speaks of 'the free for the free, the slave for 
the slave, and the female for the female*. Where Muhammad was 
not in alliance with a tribe, however, he was presumably either 
actually or potentially at war; and in war almost anything was 
permissible. For the military prestige of the ummah it was essential 
in Arabian conditions that no Muslim should go unavenged, but 
the more non-Muslims whom their tribes could not avenge, the 
greater the prestige of the ummah. The execution of B. Qurayzah 
would probably not have been regarded as: an act of war against an 
enemy by Muhammad and his contemporaries, but as the punish- 
ment of an ally which had acted treacherously. Such an action 
would, of course, have involved the ummah in a blood-feud, had 
there been any representative of Qurayzah strong enough to take 
revenge. Thus as Muhammad became stronger, there was no longer 
any question of 'normal relations' of the old type between the 
ummah and non-Muslim tribes. If they were pagan, they had a 
choice between submission to Islam and perpetual warfare; if they 
were Christian or Jewish, they might submit to Muhammad but 
retain their faith and pay tribute (jizyah). 

1 Cf. p. 32 above. 


An interesting corollary of the conception of the ummah as a new 
type of group or tribe, is that existing liabilities under the lex 
talionis were cancelled. Some such course was necessary if peace 
was to be restored in Medina. Old scores were to be wiped out. 
This is probably the point of the ruling of the Constitution ( 14) 
that 'a believer does not kill a believer because of an unbeliever', 
though it may also have had other applications. The point is 
frequently mentioned. In his letter to a 'mixed multitude'(/w;wwfl') 
in the Tihamah Muhammad promised that, if they became Mus- 
lims, any liability for blood would cease. 1 Muhammad's remark to 
al-Mughirah b. Shu'bah when he became a Muslim, that 'con- 
version cuts off what was before it j (al-isldm yajubbu md kdna 
qabla-hu), is to be taken in a similar sense. 2 And one of the points 
of the declaration to the Meccans on the submission of their city 
was that claims for usury, blood, and blood-money were cancelled. 3 
While practical needs may first have led Muhammad to formulate 
this principle, there was probably also present the idea that Islam 
was not simply a confederation of previously existing tribes and 
clans, but a new entity, and that entry into it involved a break 
with the past. 

In the relation of the ummah to other groups,, then, there are no 
great novelties. It was within the ummah that Muhammad had 
a chance to introduce reforms. What he did, when the record of it 
is read, seems to be very little, but in the circumstances of the time 
it was effective and achieved its aim of securing a large measure of 
peace within the community. 

There are certain indications that murder and other crimes were 
to be regarded as matters affecting the whole community. Thus 
the Constitution (13) states: 

The God-fearing believers are against whoever of them acts wrong- 
fully or seeks ( ? plans) an act that is unjust or treacherous or hostile or 
corrupt among the believers ; their hands are all against him, even if he 
is the son of one of them. 

Another article (21) says that, when a believer is killed (sc. by 
another believer), 'the believers are against (the murderer) entirely ; 
nothing is permissible to them except to oppose him*. With these 
regulations we may compare a verse of the Qur'an where it is 
recorded that God enjoined on the Children of Israel 'that whoever 

1 IS, 1/2. 29. 19 ( 46); translated p. 356 below. 2 Ibid. iv/2. 26. 

3 WW, 338; cf. speech at Pilgrimage of Farewell, IH, 968 ( - WW, 430). 


kills a person otherwise than (in retaliation) for another person, or 
for causing corruption in the land, shall be as if he had killed the 
people in a body (an-nas jarriCan)\ l Bell thinks that this was 
'perhaps early Medinan' and that another passage was substituted 
for it afterwards. 

On the basis of these references the hypothesis might be pro- 
pounded that Muhammad originally attempted to eliminate the 
old kinship-groups from his system of social security, but that he 
found the principle of blood-relationship so strong that he had to 
bring it back. We shall see that it was an integral part of the system 
of security finally elaborated at Medina; in the Qur'an the next- 
of-kin of a man wrongfully murdered is explicitly authorized to take 
life in revenge. 2 Though this hypothesis is attractive, however, it 
cannot be accepted, since it leaves some facts out of account. The 
early articles of the Constitution show that the old clans of Medina 
were part of the security system ; they were responsible for blood- 
money. The presumption is that this did not merely refer to rela- 
tions with groups outside the ummah, and also that there had been 
no previous way of organizing the ummah where the clans were not 
responsible for blood-money. The view that is to be preferred to 
the suggested hypothesis is that articles 13 and 21 of the Constitu- 
tion are to be interpreted negatively and positively; that is to say, 
they do not prescribe any method for the execution of justice, but 
forbid believers, on grounds of loyalty to kin, to interfere with the 
execution of justice. Since Muhammad had no police force which 
could punish offenders, the exaction of a penalty must always have 
been left to the next of kin. 

From the first, then, we conclude, the system of security at 
Medina was based on the principle of blcod-revenge by the kin- 
ship-group. Two rules, however, were stated, or restated; and the 
observance of these would prevent the cumulative tendencies of 
the blood-feud from gaining momentum and wrecking the system. 

The first of these rules was that the penalty exacted was not to 
be greater than the action for which it was a penalty. There was to 
be no question of a life for a shoe-latchet. No more than a life was 
to be taken for a life. 3 On the contrary, it was proclaimed to be 
virtuous to be satisfied with a penalty that was less than the act 

1 5- 32/35 ? E. 

2 i?- 33/35 E+ ; cf. reaffirmation of blood-ties in 8. 75/76; 33. 6. 

3 Q- 5- 45/49J 1 6. 126/127; 17. 33/35J 42- 39/37. 


penalized, or to forgive altogether. 1 On this point the Islamic ideal 
was in opposition to that of pre-Islamic Arabia. Moreover, there is 
implicit in the Islamic ideal a recognition of the equality of all 
members of the community. This is illustrated by the story of how 
in the caliphate of 'Umar a haughty scion of the desert nobility, 
Jabalah b. al-Ayham, was struck in the face by a humble member 
of the humble tribe of Muzaynah; Jabalah expected that, because 
of his importance, a severe penalty would be imposed, and when 
he was merely given the opportunity of striking the man on the 
face, he was so disgusted that he abandoned Islam and returned 
to the Christian faith and the Byzantine allegiance. 2 

The second rule was that, once revenge had been taken (up to 
the equivalent but not more), the matter was to be considered at 
an end. No penalty could justly be taken for the killing of a man 
where that was itself the exaction of a just penalty. 3 This may be 
said to be an early form of the principle that the executioner is 
not guilty of murder. The growing influence of individualism in 
the Islamic community is shown by the later view that only the 
murderer himself could justly be put to death, and by the interpre- 
tation of the Qur'an to this effect ; 4 the words of the Qur 'an in their 
obvious sense, however, permit the taking of an equivalent life 
from the murderer's tribe. This second rule was known in pre- 
Islamic times; a case is mentioned by al-Waqidi, for example, 
where Quraysh as a whole were satisfied that a quarrel was at an 
end but the dead man's family insisted on taking further revenge. 5 
Indeed, both rules are old Arab rules which are restated by the 
Qur'an. Though they appear limited in scope, they effectively 
brought internal peace to Medina. This was possible after the 
Hijrah, as it was not before, because in Muhammad there was 
a permanent arbiter to whom disputes could be referred. The 
presence of Muhammad in Medina was the novel factor in the 
situation which enabled the rules to function properly. 

The Qur'anic exhortations to forgiveness which have just been 
mentioned do not necessarily imply complete remission of the 
penalty. In a world in which blood cried out for blood, to substitute 
camels for blood 'milk for blood' as the Arabs tauntingly put it 

1 Q.5-45/49; 16. 126/127; 42. 39/37- 

2 Nicholson, Lit. Hist . 51 ; IS, i/a. 20; &c. 

3 Q. 2. 173/177 ff.J 42- 39/37; cf. 22. 60/59. 

4 Cf. J. Schacht, arts. 'Kisas' and 'atl' in El (S). * WW, 43. 


was already a measure of forgiveness, and is perhaps what is implied 
by 'forgiveness* in these passages of the Qur'an. Where a blood-wit 
was paid, there was no excuse for further shedding of blood. It is 
to be assumed that Muhammad's interest in the internal peace of 
Medina led him to encourage the acceptance of blood-money. As 
we have seen, the figure of 100 camels as the blood- wit for a man 
appears to have been known before Muhammad's time, but he 
gave official recognition both to this figure and to equivalents of it 
in other animals and in goods. 1 So far as one can tell, 100 camels or 
2,000 sheep is a high price, and the material inducement to accept 
this price must have been strong. Nevertheless, there was no 
absolute command to take money instead of blood in all cases. 
The Qur'an makes it clear that those who exact a life for a life are 
within their rights and that no judicial action can be taken against 
them. 2 If we accept the reported address of Muhammad on the 
day after the capture of Mecca, the kinsmen of a man who had been 
killed were given a choice between blood and money. 3 This may 
have been the practice of Muhammad's later years, when many no- 
madic tribes were entering Muhammad's security system for poli- 
tical rather than religious reasons; it does not easily harmonize, 
however, with a Qur'anic passage which must now be considered. 
This passage 4 distinguishes several different cases. If a believer 
deliberately kills another believer, he will be punished in Hell. If 
a believer kills another believer involuntarily or accidentally, then 
he must pay blood-money to the family of the man killed, unless, 
though the man was a Muslim, his family are unbelievers. Further, 
in all cases of involuntary killing of a believer, the person respon- 
sible, where he has the means to do so, has to pay for the freeing 
of a believing slave. This last clause is interesting, for it seems to 
be analogous to the aim of the pre-Islamic system of maintaining 
the relative strength of the two tribes involved ; in the Islamic form 
the number of believing freemen is kept constant. The fate of the 
man who deliberately kills a believer is reminiscent of the original 
inexpiability of the blood of a kinsman under the old dispensation. 5 
The distinction between deliberate and involuntary killing is 

1 Ibid. 420; cf. 338. 

* 42. 41/39; but in WW, 366 f. the acceptance of blood-money is enforced, 
perhaps because 'Uyaynah was only a non-Muslim ally. 

3 WW, 342. 4 4- 92/94 ? F. 

5 W. Robertson Smith, Kinship and Marriage in Early Arabia^ Cambridge, 
1885, 161; for an instance of commutation, cf. Lane, s.v. 'aqqa (bi 's-sahm). 


perhaps novel, but its range of application is limited, for the blood- 
wit has still to be paid. (At the present day in certain Muslim 
countries, a taxi-driver who runs over and kills a child has to pay 
blood-money, even although everyone agrees that the fault was 
entirely the child's.) In Muhammad's lifetime there were cases of 
Muslims being killed by other Muslims because the latter did not 
realize or credit the fact that they were Muslims. 1 In such cases 
the killing was deliberate, but not the killing-of-a-Muslim. The 
general impression given by the passage is that these regulations 
belong to the earlier Medinan period. In the last two or three years 
some of the details may have been allowed to lapse. There was 
much discussion in later times of how to interpret the statement 
that Hell was the punishment for the deliberate killing of a Muslim. 2 
When the passage was revealed, the community was no doubt 
small and the words interpreted literally. 

Female infanticide may be mentioned here, though to the Arabs 
it had little connexion with murder. It was an old custom among 
them (as it was in many other parts of the world) to kill a certain 
number of female children. 3 The underlying reason was poverty, 4 
coupled with the need to maintain a due balance of age and sex in 
the tribe. The strength of a tribe lay in the number of its adult 
males; though as mothers of sons the women could add to the 
strength of the tribe, the period between the birth of a girl and her 
sons' coming of age was long, and there was a limit tb the number 
of women, children, and animals whom the fighting men could 
effectively protect. Moreover girl babies seem to have been hardier 
than boys; of Muhammad's seven children by Khadljah, the three 
boys died very young, whereas the girls all reached the age of 
marriage. 5 The practice of female infanticide had also a sanction 
in the old religion. 6 The Qur'an denounces the practice as a great 
sin, and exhorts men to trust in God to provide for their needs. 7 
In the optimistic atmosphere of political and economic expansion 
there would be much less need to avoid an increase of population. 

1 Q. 4. 94/96; WW, 235 f. B. Judham; IS, iv/2. 22 f. ( = WW, 325, 366 f.)- 
'Amir b. al-A<Jbat (Ashja'). 

2 Cf. an-Nahhas, K. an-Ndsikh wa y l-Mansukh t 112 ff. 

3 Q. 81. 9; cf. Robertson Smith, 153 ff.; Roberts, 94 ff. 

4 Q. 6. 151/152 E+ ; 17. 31/33 E+ ; cf. 16. 58/60 f. c-. 

5 Cf. Ml Mecca, 38; IS, i/i. 85; &c. 

6 Q. 6. 137/138 F c E-f ; cf. sacrifice of children in the Old Testament. 

7 17- 31/33 J&c. 


Unless Muhammad proclaimed the Divine command without 
considering the economic consequences, his opposition to female 
infanticide is another indication that he had far-seeing plans for 
expansion. Unfortunately it does not seem possible to assign an 
exact date to the passages from the Qur'an. 

Apart from these rules to stop or modify criminal and anti- 
social conduct, we may be sure that in various ways Muhammad 
was constantly doing many small things that made positive contri- 
butions to the maintenance of peace and security. Where the 
solution of a difficulty was not quite clear (as when there was right 
on both sides), he would cut the knot by paying the blood-wit 
himself, 1 thus following the example of the great sayyids of the 
Jahiliyah. 2 The declaration that Medina was a sacred area (haram) 
doubtless made wrongdoers more hesitant about shedding blood 
there. 3 A severe penalty was prescribed for theft. 4 Charters were 
given defining land rights, which no doubt made it easier in the 
future to deal with land disputes. 5 

In these questions of the security of life and property, the re- 
forms introduced by Muhammad appear to be slight. Yet they 
were by no means negligible. Indeed they were fully effective, and 
for the uneasy lulls between raids to which the nomads were 
accustomed, substituted a system of social security which enabled 
the Arabs to work together for a century in the administration of 
large provinces until blood once more asserted itself over religion. 
In the conditions of Muhammad's Arabia nothing better was 
possible. He had no police force. The very idea of such a thing was 
probably unknown among the Arabs. All men were potential police- 
men or potential resisters of police. Muhammad could only have 
punished wrongdoers had he had superlative force; but for most 
of the Medinan period he had no strength to spare. In questions 
of blood, too, the old solidarity of tribe and clan was still a powerful 
force. As things were, Muhammad's combination of new and old 
appealed sufficiently to the Arabs to fire them with enthusiasm for 
the new system. Thus his treatment of these questions, without 
being revolutionary, had the revolutionary effect of achieving 

1 Cf. WW, 1 88 f. (Hassan b. Thabit), 293 f., 342 (Khuza'ah). 

2 Cf. WW, 50 f. (suggestion to 'Utbah); as-Samhudl, 146 (for political advan- 
tages B. Zurayq paid blood-wit when a Zuraqi was killed by B. Mu'alla). 

3 Constitution, 39 ; p. 224 above. 

4 Q- 5- 38/42; cf. Roberts, 90-94; EI(i). 5 IS, 1/2. 21 f. ( 19 ff.). 


(a) The existing situation 

Many facts have been recorded in Arabic literature about marri- 
age and the family in the Jahiliyah. It is difficult, however, to see 
the wood for the trees, and consequently there is no generally 
accepted view of the nature of the wood. In order to understand 
and to estimate aright the contribution in this sphere made by 
Qur'anic legislation and Muhammad's administrative practice, it 
is necessary to have some view of what existed previously. What 
follows is a tentative suggestion, though not without justification. 1 
Even if this view is at fault in many details and in much of the 
general picture, yet it enables us to make a better assessment of 
Muhammad's achievements than is otherwise possible. 

In the recorded facts about pre-Islamic Arabia there is much 
evidence that the social system was on a matrilineal basis. Thus 
we find that men and women are reckoned as belonging to their 
mother's groups. Tribes and individuals are known as sons of 
females. Property belongs communally to the matrilineal group, 
and is normally administered by the woman's uterine brother (or 
her mother's). Marriage is uxorilocal, that is, the women remain 
in their family house, and their husbands visit them. A woman 
often had several such visiting husbands, some of them probably 
concurrently. Indeed, one type of marriage, 2 or temporary union, is 
hardly distinguishable from prostitution. On the other hand, some 
of the marriages must have been relatively stable, since women 
bore half a dozen or more children to one man. 3 Divorce was also 
common, however, and many of the early Muslim women of whom 
we have biographical notices seem to have had two or more 
husbands in succession. 

While this matrilineal system was predominant over most of 
Arabia, there is also evidence of practices which have a patrilineal 
basis. The patrilineal system was strongest in Mecca, though not 
to the exclusion of all matrilineal practices, and there are also 
traces of it at Medina and elsewhere. In the patrilineal system the 
family consisted of the relatives in the male line; individuals were 
named after their fathers, and tribes were known as the sons of 

1 Grounds for holding the view and illustrative details are given in Excursus J, 
PP 373 ff- below. 

2 The fourth in al-Bukhari's tradition; cf. p. 379 below. 

3 e.g. Kabshah bint Ran' to Mu'fidh b. an-Nu'man, IS, viii. 269. 


males. Property, if communal, belonged to the patrician; if owned 
by an individual, it was inherited by his sons or consanguine 
brothers (there being no rute of primogeniture in Arabia). Marriage 
may have continued to be uxorilocal and may not have become 
virilocal, but presumably a woman had only one husband at a time. 
(It must always be remembered that the material on which we are 
dependent was written down at a time when the patrilineal system 
had superseded the matrilineal, and that the writers therefore tend 
to exaggerate the patrilineal features already present in Muham- 
mad's lifetime; thus they always give patrilineal genealogies, but 
omit information about descent in the female line, or relegate it 
to second place.) 

About the time of the Hijrah, then, matrilineal and patrilineal 
features were found in Arabian society side by side, and often inter- 
mingled. This much is fact. The explanation of this fact, which 
is to be adopted here as a working hypothesis, is that the matrilineal 
system had been prevalent in Arabia for a long period, whereas 
the appearance of the patrilineal was comparatively recent and was 
bound up with the growth of individualism. It was argued in 
Muhammad at Mecca that the Qur'an presupposes a breakdown of 
tribal solidarity and the rise of an individualistic outlook. This 
fits in well with the appearance of patrilineal features in the social 
system. Individualism means, among other things, that a man 
appropriates to his personal use what had hitherto been regarded as 
communal, though administered by him for the common good. It 
would be natural for him at the same time to become specially inter- 
ested in his own children, 1 and to want them to succeed to the wealth 
he had appropriated. In a matrilineal family the control of the family 
property would normally pass from a man to his sister's son. 

At the time of the Hijrah, then, Arabian society was in transition. 
Individualism was growing, and along with it there was a tendency 
for matrilineal features to be replaced by patrilineal. In the tran- 
sitional stage, too, there were many opportunities for unscrupulous 
men to take unfair advantage of weaker relatives and 'feather their 
own nests'. This is the background against which the Qur'anic 
reforms must be seen. 

(b) The recognition of physical paternity 

Under the matrilineal system a woman's child belonged to her 
1 Cf. Q. 68. 14, 74. 12 f., &c. 

5788 T 


family, and it was therefore comparatively unimportant to know 
who was father of a child in the physical sense. 1 Once men became 
interested in their own children, however, they would want to have 
definite knowledge of the physical paternity of their wives' children. 
The Qur'anic reform encourages this tendency. One of its central 
points is insistence on the 'iddah or tarabbus, the waiting-period after 
a woman has been widowed or divorced before she can re-marry. 
The purpose of the waiting-period, which was normally of three or 
four months, was to discover whether the woman was pregnant by 
her previous husband. In the case of divorce the man, if he was 
a 'gentleman', would do nothing during the waiting-period that 
would prevent cancellation of the divorce should his wife present 
him with a son. 2 Where a marriage had not been consummated, 
there was no need to observe a waiting-period on divorce. 3 If, 
however, a woman was pregnant, the waiting-period was until 
after the birth of the child ; after that the father, if he divorced her, 
had to provide food and clothing for her while she was suckling 
the child, and this was to be for two years unless they agreed to 
make it shorter or to give the child to a wet-nurse. 4 

(c) Plurality of wives 

It has commonly been held in Christendom that the distinctive 
feature of Islamic marriage is the permission to have four wives. 
The practice is based on a curious verse of the Qur'an (4. 3) : 

If ye fear that ye may not act with equity in regard to the orphans, 
marry such of the women as seem good to you, two or three or four 
but if ye fear that ye may not be fair (to several wives), then one (only) 
or what your right hands possess. . . . 

The interesting point is that the verse is not placing a limit on 
a previous practice of unlimited polygyny. It is not saying to men 
who had had six or ten wives * You shall not marry more than four*. 
On the contrary it is encouraging men who had had only one wife 
(or perhaps two) to marry up to four. It is not the restriction of an 
old practice but the introduction of something new. 

The verse says nothing about where the spouses are to live. 
Presumably, however, virilocal marriage was intended. There 
would have been little novelty in a man having a number of wives 

1 Cf. p. 383 below. 2 Q. 2. 226-32, 234 f.; 65. 1-4. 

3 2. 236/237 f.; 33. 49/48. 4 2. 233; 65. 6. 


whom he merely Visited*. Muhammad's own marriages were viri- 
local, though each of his wives had her own apartment. Some 
extra-Qur'anic regulations also imply virilocality. A divorced 
woman had to wait with her husband's people until her waiting- 
period was over; 1 and even in the case of a woman whose husband 
died without leaving her a house of her own or any money, the 
ruling was that she must remain where she was and not return to 
her brothers until the fourth month. 2 This evidence is late and 
might reflect later conditions, but it must always have been diffi- 
cult in an uxorilocal society for a man to have several marriages 
at once. It may be assumed, then, that the verse is primarily 
encouraging men to establish multiple virilocal families, though, 
if other arrangements could be made, it would not exclude these. 3 
European scholars have recognized that this verse of the Qur'an 
is an exhortation and not a restriction, and have further asserted 
that there are no clear cases of polygyny at Medina before Islam. 4 
Ibn Sa' d's biographies, of course, have numerous examples of men 
who had more than one wife ; but this is balanced by the examples 
of women with more than one husband. We generally have no 
information whether the marriages were contemporary or not; in 
the case of the women one husband is usually said to have * followed* 
another, but this way of putting things might be due to the environ- 
ment of Islamic practice when the facts were written down. In 
some cases it looks as if a man and woman of the same tribe married 
and lived in the same house, while the man sometimes Visited* 
a woman of another tribe, and the woman perhaps received 'visits' 
from a strange man; thus in the sub-clan of Dinar b. Malik of 
an-Najjar, of four women who had two (or in one case three) 
husbands, all had one husband from their own sub-clan of Dinar, 
one had a second husband from Dinar, and the others had second 
(and third) husbands from neighbouring clans. 5 There are a few 
examples (including some from Medina) of a man marrying two 
sisters, 6 and this may be a step towards polygyny, even though the 
marriage was still uxorilocal. Abu Uhayhah Sa'id b. al-'As (of 'Abd 
Shams of Quraysh) is said to have been the first to join together 

1 65. 6. * IS, viii. 267 f. 

3 Cf. J. Wellhausen, 'Die Ehe bei den Arabern', 469. 

4 Gertrude H. Stern, Marriage in Early Islam, London, 1939, 62, 81 ; cf. 
Snouck Hurgronje, Verspreide Geschriften, i. 233. 

5 IS, viii. 320 f., omitting Huzaylah. 

6 Cf. p. 387, n. i below. 


two sisters; 1 as this was at Mecca his household may have been 
virilocal but we cannot be certain. A man of at-Ta'if is said to have 
had ten wives, apparently at once; 2 but we really know nothing 
about the social system in which this occurred, and it would be 
rash to generalize from it. There seem, therefore, to be good 
grounds for holding that in pre-Islamic Arabia, and especially in 
Medina, it was unusual for a man to have more than one wife living 
with him in his house. 

Such a view is in harmony with the traditional account that the 
verse about plurality of wives was revealed shortly after the battle 
of Uhud. In that battle some seventy Muslims, mostly Medinans, 
were killed, so that the number of widows for whom the Islamic 
community had to care must have been considerable. It was doubt- 
less in order to meet this sudden increase in the number of 
unattached women that encouragement was given to polygyny. It 
has also to be noted, however, that the Qur'an connects the matter 
with just conduct towards Orphans' (yatama). This suggests that 
the crux of the problem of excess women was not the widows but 
the unmarried girls who now came under the guardianship of 
uncles, cousins, and other kinsmen. With some hints from the 
sources 3 we can imagine the treatment women and girls might 
receive from selfish and unsympathetic guardians ; they would be 
kept unmarried so that the guardian could have unrestricted 
control of their property, and it would be difficult for them to 
obtain legal redress against their legal protectors. The matter 
would be specially irksome in matrilineal Medina if, as seems likely, 
guardianship now went in the male line. This, then, is the situation 
the Qur'an tries to meet by encouraging polygyny. It probably did 
not intend that the guardians should themselves marry their wards, 
though, where the wards were outside the forbidden degrees, this 
would be possible. The idea seems rather to be that, if the Muslims 
generally adopt polygyny, it will be possible for all girls to be 
properly married as soon as they reach marriageable age. 

The excess of women which this practice of polygyny pre- 
supposes is sufficiently accounted for by the battle of Uhud and 
the other Muslim expeditions. It may be, however, that an excess 
of women was a regular feature of Medina and Arabia about this 

1 Ash-ShahrastSni, Kitdb al-Milal wa 'n-Nihal, ed. W. Cureton, London, 
1846, 440. a Usd t s.v. Ghaylan b. Salamah, &c. 

3 Cf. case of Kubayshah bint Ma'n, Robertson Smith, 84, 268 ff. 


time. The number of men killed in forays and the greater chances 
of survival of girl babies would tend to cause a surplus of females, 
but it would partly be offset by female infanticide and the sale of 
women captives out of the country as slaves. 1 Yet, whatever the 
position before Muhammad's time, there must have been some 
excess of women after Uhud. Moreover, the Qur'an encourages 
marriage and the procreation of children. 2 This is a corollary of 
confidence in the goodness and the success of Muhammad's move- 
ment. Because God will provide, poverty need no longer deter men 
from marriage, just as it should no longer cause them to kill female 
infants. The first successes of Islam infused confidence into the 
Muslims, though, so long as the issue was not certain, in having 
large families they were staking all on Muhammad's victory. He 
himself, with far-seeing plans for expansion beyond Arabia, may 
have been aware of the need for increased manpower. It should 
also be mentioned that the Arabs knew how to avoid conception by 
coitus interruptus ('azl), and Muhammad is said to have sanctioned 
this ; even if the reports are true, however, and they are open to 
doubt he would seem to have permitted the practice only on 
occasions when there was some special reason for it. 3 

We conclude, then, that virilocal polygyny, or the multiple viri- 
local family, which for long was the distinctive feature of Islamic 
society in the eyes of Christendom, was an innovation of Muham- 
mad's. There may have been some instances of it before his time, 
but it was not widespread, and it was particularly foreign to the 
outlook of the Medinans. It remedied some of the abuses due to the 
growth of individualism. It provided honourable marriage for 
the excess women, and checked the oppression of women by their 
guardians ; and it thereby lessened the temptation to enter into the 
loose unions allowed in the matrilineal society of Arabia. In view 
of some of the practices hitherto current, this reform must be 
regarded as an important advance in social organization. 

(d) The attitude to looser forms of union 

The Arabic word nikdh y usually translated 'marriage', is wider 
in meaning than its European equivalents. Its sense in Islamic law 
has been defined as *a contract for the legalization of intercourse 
and the procreation of children'. 4 Forms of union sanctioned by 

1 Cf. p. 270 above. z Q. 24. 32; cf. 2. 223. 

* Al-Bukhari, Nikdh (67), 96; WW, 366. 

4 A. A. A. Fyzee, Outlines of Muhammadan Law t London, 1949, 74. 


custom in pre- Islamic Arabia are called types of nikdh in Arabic, 
though in European languages some of them are nearer to prostitu- 
tion than to marriage. 1 It is not necessary here to discuss the types 
of polyandry in detail, but merely to notice that the Qur'an pre- 
supposes a distinction between those women who kept themselves 
to one man at a time and those who apparently with the full 
sanction of custom did not. These two classes of women may 
be called 'monandric and 'polyandric' respectively. Such polyandry 
was contrary to the Qur 'an's insistence on the recognition of physical 
paternity, and accordingly we find the Qur'an attempting to lessen 
polyandry and to promote monandry (in the senses indicated). 2 

Thus in 5. 5/7 the believers are given permission to contract 
marriages with women of both the Ansar and the Jews of Medina, 
subject only to the provision that the women should observe 
monandry; this was doubtless during Muhammad's first year in 
Medina. The financial difficulties of establishing monandry, in 
view of the large 'dowers' required, are shown by the exhortation 
to continence until God provides sufficient wealth (in 24. 33). 
The same verse also refers to another aspect of the financial 
problem, namely, the vested interests of those under whose control 
'polyandric' women found themselves. If these women want to 
become 'monandric', the believers are not out of greed to force 
them to engage in polyandric practices. It is impossible to say what 
precisely is involved in this last situation. 

Temporary unions with 'polyandric' women appear to have 
been sanctioned by 4. 24/28. Several early Muslim scholars took 
the verse as permitting mufah y but held that it had been abrogated. 3 
The words added in some texts, 'up to a fixed date', would confirm 
this interpretation. The practice of mut'ah is usually referred to as 
'temporary marriage' in books on Islamic law, though it is hardly 
marriage at all and the name of 'wife' is not properly applied to 
the woman contracted in mut'ah. 4 The practice is only permitted 
among a small section of the Shfah. It is a union for a fixed period, 
the period being explicitly stated in the contract. At the end of the 
period the union automatically ceases without any divorce, and 
the woman has to observe an *iddah or waiting period of only two 
courses. As part of the contract the woman receives a specified 

1 Cf. p. 379 below. 

2 For a justification of what follows cf. Excursus K. 

3 Cf. an-Nabfcas, K. an-Ndsikh wc?l-Mansukh t 105 f. 4 Cf. Fyzee, 102. 


'dower'. 1 The practice ofmut'ah according to this description is not 
the continuance of a pre-Islamic custom, but a modification of such 
a custom by Islamic ideas, notably the 'iddah. The most objection- 
able features are the possibility of complete secrecy and the fact 
that the mut'ah need last no longer than one day. Thus it would 
be easy for a woman who was attached to the old ways or who 
regarded paternity as unimportant to slip into polyandry. Why 
after a union of a day or two should she wait two months before 
accepting the advances of another suitor? It may well be, then, 
that the practice of forming temporary unions was tolerated during 
at least part of Muhammad's lifetime, perhaps even without an 
'iddah. 2 It would in any case have been difficult to stop a wide- 
spread social custom of this kind all at once. The tradition that 
c Umar I prohibited mut'ah is evidence that until his time irregular 
uxorilocal unions were formed; 3 and they may have continued 

While the Qur'an thus appears for a time to have sanctioned 
temporary unions with 'polyandric' women, the following verse, 

4. 25/29, which may be of the same date or later, facilitates the 
passage of 'polyandric' women to the practice of monandry, as well 
as making a regular marriage possible for those who had not 
enough wealth to marry a 'monandric* wife. Marriages with these 
'polyandric' women were to be made with the consent of their 
people (ahl), and the women were to observe monandry. Nothing 
is said about a limited duration; but there is the provision that, 
if the woman slipped into her old ways, her punishment was to 
be only half that of a 'monandric' woman. Thus an attempt was 
made to get Muslim men and women to abandon old Medinan prac- 
tices in so far as these involved neglect of paternity. 

As was to be expected, the attempt did not meet with complete 
success. The Qur'an shows that a number of persons must have 
clung to the old ways. With regard to them, therefore, it proposes 
a policy of segregation. A man and woman convicted of Adultery' 
(zina), which perhaps means people who have formed a union/ 
secretly and perhaps also without observing the *iddah> are to be( 

1 W. Heffening, art. 'MutV in El (i); Fyzee, 100-2; J. Schacht, The Origins 
of Muhammadan Jurisprudence, Oxford, 1950, 266 f. Nikdh as-sirr is discussed 
by I. Goldziher, 'Geheimehen bei den Arabern*, Globus, 68. 32 f. (ref. from 

5. Kohn, Die Eheschliessung im Koran, London, 1934, 83). 

2 No 'iddah according to an-Nal?fcias, 1. c. 

3 Muslim, $ahih, Nikal?, 16-18 (see El). 


flogged and forbidden to marry believers. 1 A little later in the same 
surah there comes the phrase, 'bad women to bad men and bad 
men to bad women, good women to good men and good men to 
good women'; and in this we may suppose that 'bad' is equivalent 
to 'adhering to the old customs' and 'good' the opposite. 2 By thus 
making those who held to the old ways a class apart, it was doubt- 
less hoped that in time all would adopt the new principles. 

(e) The forbidden degrees 

In Islamic law, on the basis of the Qur'anic verses 4. 22/26 f., 
a maQ jnay iiot marry his mother, his daughter, his sister (including 
half-sister, consanguine or uterine), his aunt (paternal or maternal), 
his brother's or sister's daughter, his wife's mother or daughter, 
or his father's or son's wife. Milk-relationship has a similar effect 
to blood-relationship, and marriage with foster-mother or foster- 
sister is expressly forbidden. Mother is taken to include grand- 
mother, and so on. Marriage to two sisters at the same time is also 

From the Qur'an itself we learn that until this time marriage 
with the father's wife and marriage with two sisters together had 
been practised. It has been argued that there was no bar to 
marriage in the male line, except that a man could not marry his 
daughter; 3 and this conclusion may be accepted with the proviso 
that it does not necessarily hold of all sections of Arabian society. 
On the other hand, it seems to be the case that 'on the mother's 
side all relations nearer than cousinship barred marriage'. 4 Conse- 
quently the Qur'anic law of forbidden degrees of consanguinity 
amounts to the application to the father's side of the rules already 
applicable to the mother's side. In respect of affinity, the pro- 
hibition of marriage with a step-daughter and probably also that 
with a daughter-in-law was novel, as well as that with a step- 

This account of the element of novelty in the forbidden degrees 
of the Qur'an fits in well with our general picture of the reorganiza- 
tion of marriage and the family instituted by Muhammad. One 
noticeable feature is the prominence given to paternity and patri- 
lineal descent. The rules about bars to marriage insist that blood- 

1 24. 2 f.; cf. 4. 15/19 f. 2 24. 26. 

3 Robertson Smith, 163 ff. ; with his instance of half-sister contrast IS. iii/i . 87. 

4 Ibid. 


relationship on the father's side counts equally with that on the 
mother's. Thus, without any rejection of matrilineal principles in 
this sphere, patrilineal ones are added to them, and we have 
a sort of compromise. The retention and indeed extension of 
the principle that milk-relationship is on the same level as blood- 
relationship may be regarded as a concession to matrilineal groups. 
Possibly some of those which practised forms of polyandry avoided 
undue endogamy by making certain degrees of milk-relationship 
a barrier to marriage. It is significant that it is a Meccan woman, 
'A'ishah, who finds it strange that she is allowed to appear unveiled 
before her paternal uncle by fosterage, remarking that it was a 
woman and not a man who suckled her. 1 

Another concession to the practice of matrilineal groups which 
may be mentioned here is that of khuV or * divorce by mutual 
consent'. This is probably a relic of pre-Islamic uxorilocal 
marriages where the woman or her brother had power to dismiss 
the husband. In Islamic law this power of dismissal is transformed, 
but the initiative still rests with the woman. She may ask her 
husband to divorce her, mentioning some compensation she will 
give him (such as abandoning her 'dower' to him or undertaking the 
suckling of his child); but the husband is within his rights in 
refusing to divorce her. 2 

Perhaps the most important trend to be noticed in the rules 
about forbidden degrees is that they attempt to uproot all practices 
in which the individual is not treated as an independent person. 
In this category would come the prohibition of marriage with 
a step-mother, since in certain groups in pre-Islamic times a son 
on his father's death had the right to marry the widow, apart from 
any consent on her part. Indeed, more generally, a man's heirs 
had a right to marry the women under his guardianship, including 
his wife and daughters. 3 Under the old system, where the new 
guardian was in a position to marry a woman in his care, she had 
no means of redress against him if he chose to abuse his position. 
The Qur'an shows great concern for this problem which is the 
chief one underlying its frequent references to orphans ; the be- 
lievers are not to inherit women against their will. 4 The sources do 

1 Al-Bukhari, Nikdh (67), 117; cf. Stern, op. cit. 100. 

2 Robertson Smith, 92; Stern, 1291!.; Fyzee, 139-42; G. H. Bousquet and 
L. Bercher, Le Statut Personnel en Droit Musulman Hanefite, Tunis, n.d., 1 18 ff. 

3 Robertson Smith, 86 ff. 4 Q. 4. 19/23; contrast Bell's note. 


not make it clear whether in early Islam a woman's consent was 
needed for marriage, but there are several cases in which Muham- 
mad intervened when a woman was married against her will. 1 It 
seems probable, therefore, that the prohibition of marriage with 
a niece a practice of which several pre-Islamic instances have 
been recorded 2 was directed towards increasing a woman's free- 
dom from customary restraints. Perhaps something the same was 
true of the rule forbidding marriage to two sisters simultaneously. 

Little is known about the practice of adoption in pre-Islamic 
times, and we can only guess why Muhammad stopped it. Perhaps, 
when a man married the chief woman in a household, he automatic- 
ally became 'father' of any sons and daughters living with her, and 
of any persons reckoned as sons or daughters. 3 Zayd b. Harithah 
may have become Muhammad's 'son' when Muhammad married 
Khadijah, rather than when he freed him. He was apparently 
known as Zayd b. Muhammad. In one of the verses dealing with 
Muhammad's marriage to Zayd's divorced wife, Zaynab, the 
words occur 'ascribe them to their (real) fathers ... if you do not 
know their fathers, then (let them be) your brethren in religion 
and your clients' ; these may refer to a matrilineal household where 
the physical paternity of a woman's children was known but for 
social purposes her husband was reckoned as their 'father'. 4 Among 
the meanings given for da'i (the word commonly translated 
'adopted son') are 'one who claims the relationship of a son to one 
who is not his father' and 'one who is claimed as a son by one who 
is not his father'. 5 All this suggests that we are dealing with social 
customs and not with formal acts of adoption; and these social 
customs are features of the old family organization that are undesir- 
able and to be eradicated. 

It has often been alleged that permission to marry the former wife 
of an adopted son was proclaimed only because Muhammad 
wanted to marry Zaynab. This allegation is an unjustified infer- 
ence. It is not only in this case that actual physical relationship is 
insisted on. In the verse of which a part has been quoted, the 
practice of zihar, or divorcing a wife irrevocably by swearing you 
regard her as your mother, is condemned on the grounds that this 
does not make her really your mother. More illuminating, however, 
is the rule which permits marriage with a step-daughter provided 

1 Stern, 32-36; IS, viii. 334 f., &c. 2 Stern, 62, 173 f . 

3 Cf. Robertson Smith, 112 f. 4 Q. 33. 4. 5 Lane, s.v. 


the marriage with her mother has not been consummated. 1 There 
seems to have been a general attack on fictitious or should we 
say 'merely social' ? relationships which placed restraints on the 

Some ideas about forbidden degrees may have come to Muham- 
mad from the Jews, but he differs from the Jewish practice in 
forbidding marriage with nieces. 2 Thus, while he was no doubt 
anxious that the revelation through him should be in agreement 
with previous revelations, he was also well aware of the problems 
of his own milieu. There was no blind adoption of Jewish rules for 
the sake of conformity, but those adopted were in fact appropriate 
to Medinan conditions. The similarity of the needs of Medina to 
the needs of the Israelites for whom the Levitical rules were 
written down both were settled communities with a nomadic 
background may have contributed to the similarity of the result, 
independently of deliberate imitation. 

In passing, another point may be noted at which woman is treated 
in Islam as an individual, namely, that she personally receives the 
'dower* paid by the bridegroom. ('Dower' is the usual translation 
of mahr in books on Islamic law, though the term used by anthro- 
pologists is 'bridewealth'.) The evidence for the pre-Islamic 
situation is fragmentary, and it may be that Islam merely consoli- 
dated a social trend that was already dominant. In uxorilocal 
marriages the 'dower' was sometimes given to the father or 
guardian of the bride, though there is a pre-Islamic instance of 
virilocal marriage where a gift of 'estates' was made to the bride. 3 
In Islam it is assumed rather than enacted that the 'dower' is the 
woman's. The Qur'an mentions 'dower' only incidentally in con- 
nexion with divorce; the hire (ujur) to be given to women (4. 
24/28 f.) is probably something different. Tradition tells us, too, 
that Muhammad forbade the practice of shighdr whereby two 
males or groups of males without any 'dower' exchanged daughters 
or sisters for matrimonial purposes. 4 

Little need be said about other aspects of the marriage pro- 
visions of early Islam. Concubinage with slaves or captive women 
was permitted and regulated. Divorce, which had been of frequent 

1 Q. 4 . 23/27- 

2 Robertson Smith, 166; Leviticus ', xviii; cf. Q. 24. 31. 

3 Robertson Smith, 102, quoted on p. 385 below. 

4 Al-BukhSri, Nikdh (67), 28; Robertson Smith, 91; cf. Lane, s.v. 


occurrence in pre-Islamic times (unless one prefers to say that 
most unions had been temporary), was likewise brought under 
regulation. The rule that after a man has divorced his wife three 
times he cannot take her back until she has had sexual intercourse 
with another man is possibly made in the interests of the woman, 
either to prevent husbands pronouncing divorces lightly, or to 
prevent a man in effect breaking off marital relations without 
giving the woman freedom to remarry. 

(f ) The social aspects of Muhammad's marriages 

While the personal aspect of Muhammad's marriages is best 
linked with a discussion of his character, the social aspect may 
conveniently be mentioned here, since it both illustrates what has 
been said and is illuminated by it. 1 

The most noticeable feature of Muhammad's matrimonial prac- 
tice is his establishment of a plural virilocal family. This was appar- 
ently done in the early Medinan period. At the Hijrah Muhammad 
had only one wife, Sawdah, and on the building of his residence in 
Medina called 'the mosque' in the sources she was assigned 
an apartment there. Other apartments were added for the other 
wives. It is usually assumed that 'A'ishah, the first to be married at 
Medina, went at once to her apartment in 'the mosque' ; but one 
account says that the marriage was consummated in her father's 
house. 2 In view of 'A'ishah's youth she may well have remained 
with her mother for some time ; but the housing of Muhammad's 
wives in his residence can hardly have been much later than the 
revelation of the verse about plurality of wives, probably in 625/3. 
The tradition is that Muhammad slept in the apartment of each of 
his wives in turn. 

There is some evidence that, besides his regular marriages and 
his unions with concubines, Muhammad had relations with women 
in accordance with the older matrilineal customs. The relevant 
verse of the Qur'an (33. 50/49) permits him to marry believing 
women who 'offer themselves to him*. Some seem to have done 
this, but the evidence is not clear. 3 

Another important feature is the superior status that is gradually 
given to Muhammad's wives. The first stage in marking them off 
from other women is the institution of the hijab, usually translated 

1 Cf. p. 329 below; for details see Excursus L, pp. 393 ff. 

2 Tab. i. 1263. 3 Cf. Excursus L. 


'veil', though originally it was rather a 'curtain' 1 The verse pre- 
scribing it deals also with other matters. 

O ye who have believed, do not enter the houses of the prophet . . . 
without observing when he is ready, and without announcing your- 
selves for an interview ; verily that has been insulting to the prophet. . . . 
When ye ask them (his wives) for any article, ask them from behind 
a curtain; that is purer for your hearts and for theirs. 2 

The following regulation probably belongs to the same time. 

prophet, say to thy wives, and thy daughters, and the womenfolk 
of the believers, that they let down some (part) of their mantles over 
them ( ? cover their faces) ; that is more suitable for their being recognised 
and not insulted. 3 

There are various stories giving reasons for these rules. 4 At the 
wedding-feast of Zaynab bint Jahsh some of the guests stayed too 
long and were a nuisance. At this meal or some other the hands of 
men guests touched the hands of Muhammad's wives. In the absence 
of indoor sanitation the women had to go out at night, and were 
sometimes insulted by Hypocrites; the insults may have been 
deliberate, but the perpetrators could give the excuse that they 
had mistaken Muhammad's wives for slaves. 

The fundamental reason was doubtless that with Muhammad's 
growing importance his residence was more and more a place of 
public resort. There would always be people in the courtyard round 
which were the apartments of his wives. One way in which a man 
could obtain favours from Muhammad would be to find one of the 
wives to make the request. The society of Medina looked with sus- 
picion on any private interview between a woman and a man not 
closely related to her, and consequently some protection was neces- 
sary for Muhammad's wives if scandals were to be avoided. The 
'affair of the lie', in which hostile tongues in Medina made the most 
of an unfortunate incident involving 'A'ishah, and which occurred 
shortly after the introduction of the 'veil', shows how careful 
Muhammad had to be. One of the rumours spread to discredit 
'A'ishah was that in the days before the 'veil' she had had several 
friendly conversations with the young man who rescued her. 5 

1 Cf. Stern, in ff.; Cl. Huart, art. 'Fttdjfib' in El (i); Snouck Hurgronje, 
Verspreide Geschriften, i. 309 f., &c. 

2 33- 53- 3 33- 59- 

4 IS, viii. 124 ff.; cf. N. Abbott, Aishah, 20-29, with further references. 

5 Cf. El (i), art. "A'ishah'; Abbott, 32, &c. 


Regulations prescribing modesty for all believing women were 
revealed a little later; they were to cast down their eyes, guard 
their private parts, throw their scarves over their bosoms and not 
show their ornaments except to near relatives. 1 Likewise no one 
was to enter another man's house without receiving permission. 2 
Even if this shows that the general moral level was low and needed 
to be raised, it also marks the growth of individualism in insisting 
on respect for privacy. 

A further stage in the separating of Muhammad's wives from 
other women is connected with 'the verse of the choice' (33. 28 f.), 
usually assigned to 630/9, though it may be earlier. 3 The under- 
lying reason was perhaps that Muhammad's rapidly increasing 
wealth was leading to increasing jealousy between his wives. They 
kept pestering him for clothes and articles of luxury. Zaynab bint 
Jahsh became annoyed when she thought ' A'ishah had given her 
less than her fair share of something. 'A'ishah and Hafsah were 
jealous of Mariyah the Copt. 4 Whatever the precise incident which 
led to it, there was a crisis. Muhammad withdrew from all his wives 
for a month and apparently threatened to divorce them all. At 
length he received the command : 

O prophet, say to thy wives: 'If ye desire the life of this world and 
its adornment, then come, I shall make a provision for you and send 
you forth elegantly; but if ye desire God and His messenger and the 
future abode, then God has prepared for those of you who do well 
a mighty reward.' 

This was in effect a choice between divorce and continuation of 
their marriages on any terms dictated by Muhammad. Some further 
verses indicated in general terms the sort of conduct expected of 
them (33. 30 ff.): 

wives of the prophet, whoever of you commits a manifest indecency, 
for her the punishment will be doubled twice over. . . . But to whoever 
of you is obedient to God and His messenger, and acts uprightly, We 
shall give her reward twice over, and We have prepared for her a noble 
provision. O wives of the prophet, ye are not like any ordinary woman ; 
if ye show piety, do not wheedle in your speech, so that one in whose 
heart is disease grow lustful, but speak in reputable fashion. Remain 
in your houses (or 'sit with dignity*), and do not swagger about in the 
manner of the former paganism. . . . 

1 24. 32/31. 2 24. 28/27 f. 3 IS, viii. 129 ff.; cf. Abbott, 49-59. 
4 Cf. Q. 66. 1-5 and commentaries; Abbott, 50, 59. 


c A'ishah and eight other wives are said to have chosen God and His 
messenger. Perhaps it was on this occasion that Muhammad 
divorced some of the women classed as * married to Muhammad 
and divorced* ; a wife from the tribe of 'Amir is said to have chosen 
to be sent away, but her identity is uncertain. 1 

From now on Muhammad's wives had an honoured and impor- 
tant place in the community. He had probably no intention of 
imitating the monarchs of Persia and other oriental countries, who 
increased their own dignity by special arrangements for their wives. 
Nevertheless the regulations did have an effect of this kind. The 
believers were forbidden to marry Muhammad's wives after him. 2 
Permission to do so would have increased the disunity in the com- 
munity. Perhaps it was by way of compensation for this restriction 
that the wives came to be known as 'mothers of the believers'. 

The last feature to be noted about Muhammad's marriages is 
that he used both his own and those of the closest Companions to 
further political ends. This was doubtless a continuation of older 
Arabian practice. All Muhammad's own marriages can be seen to 
have a tendency to promote friendly relations in the political sphere. 
Khadljah brought him wealth, and the beginnings of influence in 
Meccan politics. In the case of Sawdah, whom he married at Mecca, 
the chief aim may have been to provide for the widow of a faithful 
Muslim, as also in the later marriage with Zaynab bint Khuzay- 
mah; but Sawdah's husband was the brother of a man whom 
Muhammad perhaps wanted to keep from becoming an extreme 
opponent; 3 and Zaynab's husband belonged to the clan of al- 
Muttalib, for which Muhammad had a special responsibility, while 
he was also cultivating good relations with her own tribe of 'Amir 
b. Sa'sa'ah. His first wives at Medina, 'A'ishah and Hafsah, were 
the daughters of the men on whom he leaned most, Abu Bakr and 
'Umar; and 'Umar also married Muhammad's grand- daughter, 
Umm Kulthum bint 'All. Umm Salamah was not merely a deserv- 
ing widow, but a close relative of the leading man of the Meccan 
clan of Makhzum. Juwayriyah was the daughter of the chief of the 
tribe of al-Mustaliq, with whom Muhammad had been having 
special trouble. Zaynab bint Jahsh, besides being Muhammad's 
cousin, was a confederate of the Meccan clan of 'Abd Shams, but 
a social motive may have outweighed the political one in her case 

1 IS, viii. 138. 2 Q. 33. 53. 

3 Suhayl b. 'Amr; cf. M/Mecca t 140; also pp. 56-64 above. 


to demonstrate that Muhammad had broken with old taboos. 
Nevertheless the clan of 'Abd Shams, and Abu Sufyan b. Harb in 
particular, were in his thoughts, for Abu Sufyan had a Muslim 
daughter, Umm Habibah, married to a brother of Zaynab bint 
Jahsh; and when the husband died in Abyssinia, Muhammad sent 
a messenger there to arrange a marriage with her. The marriage 
with Maymunah would similarly help to cement relations with her 
brother-in-law, Muhammad's uncle, al-'Abbas. There may also 
have been political motives in the unions with the Jewesses, Saf lyah 
and Rayhanah. In so far as there are any solid grounds in the 
accounts of his marriages or proposed, marriages with women 
in the 'supplementary list', the dominant motive was presumably 
political. They nearly all came from nomadic tribes or places 
at a distance. 

It is noteworthy that Muhammad had no Medinan wife. Layla 
bint al-Khatlm (Zafar) is said herself to have arranged a marriage 
with Muhammad, but to have been forced by her people (qawiri) 
to give up the project. 1 Likewise Muhammad is said to have 
thought of marrying Habibah bint Sahl (Malik b. an-Najjar), but 
to have refrained because of the Ansar. 2 Clearly he could only be 
successful in Medina if he was impartial, and his impartiality 
would be seriously infringed by such marriages. Abu Bakr married 
a woman of the Khazraj, apparently towards the end of his life 
(the uterine sister of Sa c d b. ar-Rabf), and 'Umar h^d a wife from 
the Aws. On the whole, however, there was very little inter- 
marriage at Medina between the Meccans and the Medinans, 
perhaps because of the differences in the social systems. 

Other two important Companions, 'All and 'Uthman b. 'Affan, 
were bound to Muhammad by marriages with his daughters, 
Fatimah and Ruqayyah (followed by Umm Kulthum); 'All also 
married Muhammad's grand-daughter (by Zaynab), Umamah bint 
Abi 'l-'As. 3 Az-Zubayr b. al-'Awwam was married to Abu Bakr's 
daughter Asma'. 'Abd ar-Rahman b. 'Awf, on being sent in com- 
mand of an expedition to Dumat al-Jandal in 627/6, was told to 
marry the daughter of the chief if he submitted. Thus it was by 
no means only Muhammad's own marriages that were political, 
though in his case, as head of the community, there were special 
reasons for taking political considerations into account. 

1 IS, viii. 107 f. 

2 Ibid. 326 f. 3 Ibid. 169. 


(g) Conclusion 

In the sphere of marriage and family relations Muhammad 
effected a profound and far-reaching reorganization of the structure 
of society. Before his time new individualistic tendencies were 
certainly present, but their presence was leading more to the 
breakdown of the old structure than to the building up of a new 
one. Muhammad's essential work here was to use these individual- 
istic tendencies in the raising of a new structure. The customs and 
practices of the communal (tribal) stage of society, to vary the 
metaphor, had suffered shipwreck; Muhammad salvaged what was' 
valuable from them, and carried it over to the new individualistic! 
society. In this way he produced a family structure that in many 
respects has proved attractive and satisfactory for societies emerg- 
ing from the communal stage and passing into an individualistic 

Both by European Christian standards and by those of Islam, 
many of the old practices were immoral, and Muhammad's re- 
organization was therefore a moral advance. The old nomadic 
system may have been satisfactory in desert circumstances so long 
as it remained intact. Once disintegration commenced, however, 
it became unsatisfactory and had to go. It is to Muhammad's 
credit that he produced a viable substitute. 


A few verses of the Qur'an 1 state succinctly but in some detail 
the rules for the division of an inheritance on a man's death, and 
these have been elaborated into a complex system by later jurists. 
It would be confusing rather than illuminating to discuss the 
matter fully here, but it is important to try to understand the 
fundamental principles. 2 It will be found that Muhammad's 
enactments are aimed at eliminating the abuses which arose in the 
change from a communal system of ownership to an individualistic 

There is much that is obscure about the existing practice in 
regard to inheritance at the time of the Hijrah, but certain general 
features are clear. In Medina, where society was mainly matrilineal, 

1 Q. 4. 11/12-14/18, 176/175- 

2 Cf. S. Vesey- Fitzgerald, Muhammadan Law, London, 1931, in ff.; Fyzee, 
Outlines of Muhammadan Law, 331 ff.; F. Peltier and G.-H. Bousquet, Les 
Successions Agnatiques Mitigtes, Paris, 1935. 

5783 U 


a woman could not own property; this possibly only means 
that under the communal system it was administered by her 
maternal uncle, uterine brother, or son; when an administrator 
died, he would be succeeded by the next most eligible person in 
the matrilineal group. If there were patrilineal groups where 
property was held in common, then something similar would hold 
good; a man would be succeeded by one of his brothers or sons, 
who would administer on behalf of the group ; if the group was 
large, the property might be divided between several adminis- 
trators; and the same might happen if several 'heirs' were of 
approximately equal standing and ability. Where effective adminis- 
tration was the chief consideration, there could be no question of 
considering as * heirs' anyone who was not living and above the 
age of puberty. 

With the infiltration of individualistic ideas it was easy to pass 
imperceptibly from administration to individual ownership. In 
practice this meant that the strong took everything and the weak 
had nothing. When Muhammad's grandfather died, Muhammad's 
father was already dead, and Muhammad was a minor; conse- 
quently he received nothing. The dead had no share, and Muham- 
mad was too young to share in his own right along with his uncles. 
This is doubtless part of the reason for the Qur'an's insisting on 
good treatment for orphans. But the principle that the dead do not 
inherit and that the living cannot represent the dead must have 
been deeply rooted, for the Qur'an does not attempt to make any 
change on this point. 

This is the situation with which the Qur'an had to deal. The 
new tendencies and the corresponding abuses had already appeared. 
The Qur'an accepts the tendencies and sets out to remedy the 
abuses. It does not state that property is to belong to individuals, 
but assumes that it does in fact so belong. Further, it assumes that 
property may belong to women as well as to men. The case of 
Khadijah, despite its obscurities, shows that this had been the 
practice in Mecca. In Medina, however, it was apparently a novelty, 
and there may have been some conservative opposition at first. 1 

The main aim of the Qur'anic rules was to ensure that no relative 
towards whom a man had some obligations was defrauded of his 
fair share of the inheritance. Consequently they prescribe that, 

1 Cf. Q. 4. 127/126, 'the female orphans to whom ye do not give what is 
prescribed for them*. 

viii. 3 INHERITANCE 291 

before the inheritance is handed over to the normal or agnatic heirs 
('asabat often misleadingly translated in English as 'residuaries'), 
fixed shares are to be given to certain persons in certain circum- 
stances. These persons are known as 'sharers' (ashdb al-fard'id, 
&c.) or 'Qur'anic heirs'. The main 'sharers' are the widower or 
widow(s), the parents, the daughters, and in certain cases a son's 
daughters, and the man's sisters and uterine brothers. After the 
prescribed shares have been paid, the 'residue' (normally the main 
part of the estate) goes to the sons, father, and brothers, in that 
order. For the sake of completeness the essential part of the 
Qur'anic rules may be quoted : 

In regard to your children God charges you (as follows): The male 
receives the portion of two females ; if they be women, more than two, 
then they receive two-thirds of what a man has left, but if they be only 
one she receives a half. His parents receive each of them a sixth of what 
he has left, if he have children; but if he have no children and his 
parents heir him, then his mother receives a third ; if, however, he have 
brothers, his mother receives a sixth (this) after any bequests he may 
have made or debts (have been paid). . . . 

A half of what your wives leave belongs to you if they have no 
children ; if they have children, a fourth of what they leave belongs to 
you. ... To them belongs a fourth of what ye leave, if ye have no 
children ; if ye have children, an eighth. . . . 

If a man or a woman whose property falls to be inherited have no 
direct heirs (sc. agnates), but have a (sc. uterine) brother or sister, each of 
the two receives a sixth; if there be more than that, they share in the 
third. . . . 

If a man perishes and has no children but a sister, the half of what 
he leaves belongs to her ; and he is her heir if she have no children ; if 
there be two (sisters), the two-thirds of what he leaves belongs to them ; 
if there be brothers and sisters, a share equal to the portion of two 
females belongs to the male. 

Instead of showing how these rules are applied in Islamic law, 
it will be more useful to give some examples of the distribution of 
property in the type of case which frequently occurs. 

Wife, son: receive respectively 1/8, 7/8. 

Wife, son, daughter: 1/8, 7/12, 7/24. 

Wife, two sons, two daughters: 1/8, 7/24 (2), 7/48 (2). 

Husband, two sons, two daughters: 1/4, 1/4(2), 1/8(2). 

Two daughters, father or distant agnates: 1/3 (2), 1/3. 

Two daughters, father, mother: 1/3 (2), 1/6, 1/6. 


Father, mother: 2/3, 1/3. 

Father, mother, brother: 5/6, 1/6, nil. 

Father, mother, wife, two sons, two daughters: 1/6, 1/6, 1/8, 

13/72 (2), 13/144 (*) 

Husband, son: 1/4, 3/4. 

Husband, father: 1/2, 1/2. 

Father's father, two brothers: 2/3, 1/6 (2). 

Sister, no children: 1/2. 

Brother, sister, no children: 1/3, 1/6. 

There are certain further points which may be noticed. The 
system is fundamentally patrilineal. The normal heirs are the sons, 
father, and brothers. A daughter's sons do not inherit, since they 
are not members of the patrilineal clan. Even if there are no sons 
and father to be considered, a man's daughters do not get more 
than two-thirds of his estate; the remainder goes to more distant 
agnates. Nevertheless there seems to be a concession to matrilineal 
practice in the provision that, where a man has no direct heirs 
(usually taken to be child, son's child, father or father's father), his 
uterine brothers and sisters (if more than one) inherit a third 
between them. It is also significant that in this case males and 
females have equal shares, whereas males mostly have twice the 
portion of a female. (The above statements are based primarily 
on the interpretation of the Qur'an according to the Hanaf I type 
of Sunni law; there are slight differences in other types of Islamic 
law; e.g. relations through females fare better in the law of the 
Ithna 'Ashari branch of the Shi f ah, the underlying principle appar- 
ently being that relations through females and relations through 
males were equally close to a man.) 1 

The effect of the Qur'anic rules was to subdivide property within 
the simple family, and also occasionally within a slightly wider 
family group. The system to which the rules lead bears the marks 
of its origin in an environment of caravan-city and desert. It is easy 
to subdivide a herd of camels or sheep, or a quantity of merchandise 
which can be valued in money. The subdivision of land in this 
way is not so satisfactory, and tends to retard agricultural improve- 
ments. At the same time, the rules of inheritance and the sub- 
division they bring about show that individualism had by no means 
driven out all communal ideas from the outlook of the Arabs. The 
individual may have his precise share in the property of his father 
1 Cf. Fyzee, 381-406, esp. 403 ff. 

viii. 3 INHERITANCE 293 

or brother, but the family has a certain claim on the property of 
every member of it. This is seen, not only in the precise fixing of 
shares, but in the fact that the right to make bequests outside the 
family was restricted to a third of the estate, 1 while no bequests 
were allowed to persons who were heirs in any case. 2 These two 
rules are based on Tradition and not on the Qur'an, but they are in 
accordance with the spirit of the latter. They give expression to the 
conception that, though a man is owner of his property during his 
lifetime and may dispose of it as he pleases, he is also in a sense 
a steward of it on behalf of his family. Thus the Qur'an goes far 
to meet the individualism of the times, and yet is not completely 


The topics that have been dealt with so far, social security, 
marriage, and inheritance, are the only ones on which Muhammad 
carried out extensive reforms. There are a few minor matters, how- 
ever, which ought to be mentioned. 

(a) Slavery 

The attitude of the Qur'an to slavery is not unlike that of the 
New Testament. Both accept the fact of slavery and do something 
to mitigate it. The commonest source of slavery in pre-Islamic 
times was presumably the warfare between the Arab tribes. 3 In 
such raiding and fighting women and children were often carried 
off. Where their tribe could afford it, they would probably be 
ransomed ; but frequently they were sold as slaves. Zayd b. Harithah 
was thus carried off as a stripling and sold at 'Ukaz. 4 Out of the; 
86 Emigrants named by Ibn Sa f d as fighting at Badr at least another \ 
10 were freedmen or slaves ; there were also 4 freedmen and perhaps! 
i slave among the Ansar. 5 Of most of the Emigrants we are told 
either that they were captured when mere boys or else that they 
were born in slavery (muwallad). Most of them were of Arab 
descent, but there were at least one Persian and two Abyssinians, 
these foreigners being apparently born in slavery. There were 
also slaves fighting for the Meccan pagans, but the hypothesis 
that Meccan military strength rested on Abyssinian slaves is 

1 Fyzee, 71. 2 Ibid. 72. 

3 Roberts, Social Laws 1 53 ff. 4 IS, iii/i. 26. 

5 See Excursus C, p. 344 below. 


unwarranted and to be rejected. 1 If the Emigrants are a fair 
sample, the majority of slaves were Arab and not foreign. 2 

There was no objection in principle to the selling of adult males, 
as is shown by the sale of Muslim prisoners by B. Lihyan to the 
Meccans in 625/4. I n practice, however, it would be difficult use- 
fully to retain an adult male who had been captured, since he would 
presumably try to escape whenever there was an opportunity. A 
man who had been born into slavery, on the other hand, would 
normally have no tribe to which to flee. There were obvious advan- 
tages in removing slaves far from their original region. On the 
whole, however, slaves seem to have been well treated. Despite 
their inferior status they had a recognized position in the family 
and clan, and shared to a large extent in its good fortune and bad 
fortune. Muhammad's slave, Salih Shuqran, fought at Badr for 
the Muslims while still a slave, and there are said to have been two 
other slaves among the Muslims. 3 The freedmen (mawdlt) also 
stuck closely to their patrons, on whom they were dependent for 
protection according to the pre-Islamic security system. Zayd b. 
Harithah, after receiving his freedom, chose to remain with Mu- 
hammad rather than return to his own family. 4 

The inferior status of the slave did not prevent his becoming 
a Muslim. A few did so in the early days. Abu Bakr bought some 
and freed them; and it was always regarded as a pious act to free 
a slave. 5 The freeing of a believing slave was prescribed by way of 
compensation to the community in cases where one believer had 
killed another unintentionally. 6 Provision is also made in the 
Qur'an for the manumission of slaves by a method which presum- 
ably had pre-Islamic antecedents. 7 A contract is made between the 
knaster and the slave that the slave is to pay a certain sum for his 
freedom, and the slave was able, while still a slave, to earn money 
for this purpose. It was not compulsory, however, to set even 
believing slaves free, as is shown by the presence of Muhammad's 
slave Salih Shuqran at Badr. Muhammad's concubine Mariyah, a 
Christian, was apparently not set free; and, to judge from Ibn 
Sa'd's accounts of the Badr fighters, the same was true of many of 

1 Ml Mecca, 154-7. 

2 Contrast R. Brunschvig, art. "Abd* in El (2), apparently following H. 

3 IS, iii/i. 34. 4 Ibid. 28. 5 Cf. Q. 2. 177/172; 90. 13. 

6 Q. 4. 92/94; for perjury, 5. 89/91 ; 58. 3/4. 

7 24. 33; cf. Lane, s.v. kdtaba. 


their concubines. On the contrary, the rules about the marriage of 
slaves in the Qur'an show that slavery was regarded as continuing. 1 
In 4. 36/40 they are one of several classes of dependent, weak, or 
needy persons to whom kindness is to be shown. 
v/The critics may say that, in view of his political power towards 
the end of his life, Muhammad could have done more to alleviate 
the lot of the slaves. Such a criticism rests on a false appreciation 
of the situation in which he found himself. There were many 
things which urgently required to be set right but this was not one 
of them. On the whole the slaves were not too badly treated. The 
chief disability in being a slave was that one could not of one's own 
will leave the group to which one was attached. In the Arabia of 
the early seventh century, however, this was much less of a dis- 
advantage than it would be in a more individualistic societyj 
Though the connexion of Islam with the rise of individualism has 
been emphasized throughout this study of the life of Muhammad, 
it should also be realized that individualism was only at its begin- 
nings. The family and the clan still counted far more than they do 
in Western Europe in the twentieth century. For protection and 
even livelihood the ordinary man or woman was dependent on the 
group to which he or she belonged. A strong man might break 
away from his group to the extent of making himself head of a 
sub-group. For the ordinary individual, however, the question of 
leaving the group could hardly arise. Even when a woman married 
and went to live with her husband's family, she was still to a large 
extent dependent on her own family JjWhen it is remembered that 
the slaves were either women or men born in slavery or men taken 
away from their kinship group when young, it is clear that in their 
inability to leave the group they differed from other dependent 
members of the group only in a slight degree. Freedom would only 
be valuable for them if it meant attachment to a group in which they 
would have more privileges, or an increase in privilege within the 
group in which they were slaves. In practice it was usually the latter 
which occurred at manumission. The slave did not leave the group 
to which he belonged as slave, but certain relationships within it 
were changed; the head of the group, instead of being his 'master' 
became his 'patron*. 

Moreover there are two ways in which Muhammad may be said 

1 24. 32; cf. 2. 221/220; but for \hefataydt of 4. 25/29 and 24. 33 cf. Excursus 
K, p. 391 below. 


to have done something to improve the position in Arabia with 
regard to slavery. The institution was deeply rooted in the military 
customs of the time and region. When a defeated tribe was too 
weak to retaliate or even to ransom its captives, the victors would 
kill the men and sell the women and children into slavery. The 
treatment of the Jewish clan of Qurayzah by the Muslims was 
simply the regular Arab practice, but on a larger scale than usual, 
since the Muslims were stronger than even the average strong tribe. 
Nevertheless the extension of the Pax Islamica, by reducing warfare 
and raiding, reduced the opportunities for making slaves. Indeed 
it became impossible for a Muslim to make a slave of another 
Muslim. This was an implication of the conception of the ummah, 
and it is explicitly stated in Tradition. 1 In the second place, this 
effective reduction of slavery in Arabia through Muhammad's 
activity was supported by the conception, implicit in the Qur'an 
and in many of his sayings, and sometimes also explicit, that all 
Muslims are brothers. The recognition that inequalities between 
men belong to the nature of things has been something of a counter- 
poise. 2 Nevertheless the idea of 'brotherhood' has been a powerful 
one in Islam and has aided the movement for the mitigation and 
abolition of slavery. 

(b) Usury 

It might be thought that the prohibition of usury (riba) in Islam 
was due to the wrong attitude to wealth among the rich merchants 
of Mecca. A careful examination of the Qur'an, however, makes 
it clear that this is not so, but that the prohibition was first made 
in the early years at Medina and was directed primarily against the 
Jews. 3 Richard Bell's dating of the passages referring to usury is 
unfortunately not so precise as one would desire. None is Meccan 
in his view; and for the other limit his statements might be taken 
to imply that none is later than about the siege of Medina. The 
passage which clinches the matter is the following : 

So for wrong-doing on the part of those who have judaized We have 
made (certain) good things forbidden to them which had (formerly) 
been allowable for them, . . . and for their taking usury though they had 
been forbidden to do so. 4 

1 Cf. A. J. Wensinck, A Handbook of Early Muhammadan Tradition, Leiden, 
1927, s.v. 'Slave*. * Cf. Q. 16. 75/77. 

3 Cf. J. Schacht, art. 'Riba' in El (i). 4 4. 160/158 f. F- ? G. 


There may be a reference to usury in another passage about the 
Jews. 1 Once this connexion between usury and the Jews has been 
established, it is natural to regard the threat of war against believers 
who take usury as directed against Jews. 2 The remaining passages 
suggest that those who take usury are in danger of Hell or assert 
that usury leads to no increase from God whereas zakdt does. 3 

A reconstruction of the situation would be somewhat as follows. 
In his first years at Medina Muhammad was nominally in alliance 
with the Jews. In course of time he had to appeal for contributions, 
either to support the poorer Emigrants until booty began to come 
in, or as is more likely for military preparations, especially in the 
period between Badr and the siege of Medina when Meccan attacks 
were expected. This appeal was made to the Jews as well as to the 
Muslims. Most of the Jews refused, but said they were ready to 
lend money at interest. 4 Muhammad, however, came to realize 
that it was contrary to the Jewish law to lend money at interest to 
a co-religionist. In his eyes Jews and Muslims were co-religionists, 
and therefore the Jews ought to make outright contributions to his 
cause, or at least to lend money without interest. In this way the 
question of usury becomes an aspect of his quarrel with the Jews 
about recognition of his prophethood. 

Though the prohibition of usury was directed against the Jews 
in the first place, some Arabs also may have been involved at 
Medina. The later development of both practice and theory, how- 
ever, is very obscure. 5 There was much discussion of the precise 
meaning of ribd and of what transactions were prohibited, for the 
general idea in the word is roughly 'getting more than you give*. 
The nature of the jurists' discussions suggests that ribd was com- 
paratively limited in Muhammad's time. He is said to have men- 
tioned the point in his proclamation after the conquest of Mecca; 6 
and in a letter to B. Juhaynah he specifies that they are to abandon 
the interest on sums owing to them and claim only the capital. 7 
On the other hand, there is no evidence of any attempt by Muham- 
mad to stop commercial dealings at Mecca. The caravan trade of 
Mecca and Medina continued for some time, and in the end 

1 5. 62/67 ? FG. 2 2. 278-81 ? FG. 

3 2. 275/276 f. E , E-f ; 3- 130/125 f. GH; 30. 39/38 ? F. 

4 Cf . WW, 1 64 80 dinars lent by a Jew to one of the Ansar for a year at 50 per 
cent, interest; al-Bukhari Buyu* (34), 14 Muhammad bought grain from a Jew 
and had to give a coat of mail as a pledge till he paid for it. 

5 Cf. Schacht, I.e. 6 IH, 821 ; WW, 338. 7 IS i/2. 25. 3. 


probably died a natural death. The energies of the entrepreneurs 
were absorbed in administering the conquests and making fortunes 
elsewhere; and the occupation of 'Iraq and Syria brought an easier 
route from the Indies under Muslim control. Perhaps Muhammad 
himself only tried to stop lending for consumption as distinct from 
lending for productive purposes. The later elaborations which in 
the modern age have hindered the financing of productive com- 
mercial and industrial enterprises are possibly the work of theoreti- 
cians remote from any thriving commerce. 

This is no place, however, to discuss these later developments. 
In Muhammad himself there seems to have been no intention 
of hindering legitimate trade or of revolutionizing the financial 
practices of Mecca, despite the Qur'anic criticisms of the pagan 
Meccans* attitude to money. The idea underlying the prohibition 
of usury was that all believers were brothers and therefore ought 
to help one another financially as well as in other ways. 

There is hardly anything that could be called reform in the 
other rules of the Qur'an dealing with commerce and finance. They 
exhort to upright dealing, and in certain cases prescribe that the 
matter should be put in writing presumably an innovation. 

(c) Wine-drinking 

The prohibition of intoxicating drinks is one of the well-known 
features of Islamic civilization, and has its basis in' certain verses 
of the Qur'an : 

They will ask thee about wine and may sir ; say: 'In both of them 
there is great guilt, and also uses for the people, but their guilt is greater 
than their usefulness.' 

O ye who have believed, wine, maysir, stone altars (or images), and 
divining arrows are simply an abomination, some of Satan's work; so 
avoid it, mayhap ye will prosper. Satan simply wishes to cause enmity 
and hatred to fall out amongst you in the matter of wine and maysir, 
and to turn you away from the remembrance of God and from the 
Worship, so are ye going to refrain? 1 

At the conquest of Mecca Muhammad is said to have refused a 
present of wine, and to have had the wine poured out. 2 

The only point to be discussed is that of the reason for this 
prohibition of wine-drinking. The tenor of the Qur'anic passages, 
especially if, with later Muslims, maysir is taken to include all 

1 2. 219/216; 5. 90/92 f. 

2 WW, 348- 


forms of gambling, suggests that the attitude of the early Muslim 
community was not unlike that in certain pietistic circles in Europe 
today. Some other facts support this suggestion. There is an 
ascetic strain in the Semitic temperament, and even before Muham- 
mad began to preach there were men in Mecca, like 'Uthman b. 
Maz'un, 1 who avoided wine. There are also stories of the unpleasant 
effects of drunkenness, even among prominent Companions. 2 

On the other hand, there are reasons for thinking that such a 
view is not the whole truth of the matter. The meaning of khamr 
was much discussed by later jurists, and we cannot be sure whether 
it originally meant any intoxicating drink or wine in the strict sense 
(the fermented juice of the grape). If it meant the latter, then 
political considerations may have come in, for grape-wine was 
normally imported from Syria and 'Iraq. Thus wine-drinking 
would imply trading with the enemy. This point, however, is not 
so weighty as another, namely, the connexion of wine with may sir 
in the verses quoted from the Qur'an. May sir was a practice by which 
ten men bought a camel, slaughtered it, and then drew lots for the 
portions by means of arrows ; three arrows had no portions assigned 
to them, and the men to whom these fell had to pay for the whole 
camel. The Qur'anic objection to the passage is presumably not 
that it was a form of gambling, but that it was closely connected 
with the pagan religion, since the arrows were kept by the guardian 
of the Ka e bah at Mecca. 3 It seems likely, therefore, that the main 
reason for the prohibition of wine may have been some connexion 
with pagan religion of which we are not aware. 4 

(d) The calendar 

The abolition of intercalary months is a slight change introduced 
under Muhammad which has given a definite stamp to Islamic 
civilization. The pre-Islamic Arabs observed the lunar months, 
but kept their calendar in line with the solar year by introducing 
intercalary months where necessary. The matter is referred to in a 
passage of the Qur'an : 

Twelve is the number of the months with God, (written) in God's 
Book on the day when He created the heavens and the earth; of these 

1 IS, iii/i. 286. 

2 Al-Bukhari, Sharb (42), 13; for further references cf. A. J. Wensinck, art. 
'Khamr' in El (S). 3 Cf. B. Carra de Vaux, art. 'Maisir* in El (S). 

4 Cf. J. Wellhausen, Reste Arabischen Heidentums, 2nd ed., Berlin, 1897, 114. 


four are sacred ; that is the eternal religion ; so do not wrong each other 
in them ; but fight the polytheists continuously, as they fight you con- 
tinuously, and know that God is with those who act piously. The post- 
ponement (the intercalary month nasl) is simply an increase of 
unbelief, in which those who have disbelieved go astray ; they make it free 
(not-sacred) one year and sacred another, that they may make adaptable 
the number of what God hath made sacred, and may make free (not- 
sacred) what God hath made sacred. . . .' 

Muhammad is said to have made public these verses during the 
address he gave during the Pilgrimage of Farewell. 2 

There are so many obscurities in the whole question of the inter- 
calary month that it is difficult to say what were the underlying 
reasons for the adoption of a lunar year. 3 The Qur'an implies that 
intercalation was in some respect a human activity infringing God's 
law, and contrasts the fixity of the latter with variability of the 
human device. This makes it almost certain that, despite some of 
the accounts, the Arabs had no fixed system of intercalation. As 
reason for the prohibition of intercalation there are two main 
possibilities. The method of settling when a month was to be 
intercalated may have been connected with paganism in some way 
of which we are not aware; it was certainly linked with the obser- 
vance of the sacred months. 4 Or else there may have been a risk 
that the uncertainty about which months were sacred would cause 
disputes and endanger the Pax Islamica. Whatever the reason for 
it, this adoption of the lunar year shows again the non-agrarian 
character of Islam; Islam is often said to mould or influence every 
department of life, but it has not penetrated the agricultural life 
of the millions of peasants who are good Muslims. Their farming 
practices and some of the religious ideas connected with them 
continue in the traditional way regardless of Islam. A work like the 
Georgics is inconceivable in any Islamic literature. 


The prohibition of usury, wine, and intercalated months has 
done much to give Islamic countries the appearance they present 
to the traveller; but the other matters dealt with in this chapter 

1 9. 36 f. 2 IH, 968 f.; WW, 430 f. 

3 Cf. Caetani, i. 356 ff.; Buhl, Muhammed, 350 f.; M. Plessner, art. 'Mu- 
harram' in El (i),.with further references. 

4 Cf. p. 8 above. Contrast H. Winckler in Arabisch-Semitisch-Orientalisch, 
85-90 (Berlin, Mitteilungen der Vorderasiatischen Gesellschaft, 1901, 4). 

vni. 5 CONCLUSION 301 

are more important in a consideration of the reform of the social 
structure. It remains to say a word about the relation of these 
reforms to the social aspect of the malaise of the times. 

The root of the social troubles of the Hijaz in the early seventh 
century A.D. was that the communal (tribal) system of the desert 
was breaking down in the settled life of Mecca and Medina. The 
precise reasons were different in the two places. In Mecca a mer- 
cantile economy had fostered the growth of individualism. In 
Medina the autonomy of each tribe and clan, appropriate to desert 
conditions, led in the confined space of an oasis to an insecurity of 
life that had become intolerable. Individualism meant that the 
strong oppressed the weak and neglected their traditional duties to 
clan and family. It was present at Medina, but not so noticeable 
there as at Mecca. The characteristic of the social structure of 
Medina was a tendency towards the formation of larger groupings, 
either as alliances or on the basis of kinship, real or artificial. In all 
this social disintegration most individuals were doubtless painfully 
aware of their insecurity and isolation. 

Among the nomadic tribes there does not appear to have been 
any social crisis apart from the menace to their autonomy arising 
from Muhammad's successes. There were, of course, the perennial 
problems of nomadic life constant raiding and blood-feuds and 
the recurring risk of famine. 

Against this background it is interesting to see how individualism 
and communalism were combined by Muhammad. The ummah 
or new community of Islam has as its first aim the preservation of 
peace between its members, and Muhammad, as executive head 
of the community, had to see to it that this aim was realized. But 
the ummah was much more than a method of preserving peace. In 
one respect it was a community of individuals, for Islam accepted 
the tendency towards individualism, and even encouraged it (as in 
the new family structure). The ultimate moral sanction in Islam, 
punishment in Hell, applies to the individual for his conduct as 
an individual. On the other hand, the individual was taken out of 
his isolation and insecurity and made to feel that he belonged to the 
ummah. The early practice of 'brothering' may seem artificial, but 
the sense of brotherhood between Muslims has become very deep ; 
witness such a title in our own days as 'the Muslim Brotherhood* 
(al-Ikhwdn al-Muslimin literally 'the brothers, the Muslims'). 
The ummah was a closely-knit community, thought of on the lines 


of a tribe, and much of the old mystique attaching to the kinship 
group has become attached to it. 

At the same time, however, these kinship groups continued to 
play a part in the structure of Islamic society. The ummah was 
essentially a body of clans and tribes in alliance. These still had 
a part to play in maintaining the security of life and property. In 
the regulations for inheritance, too, the claims of a man's family 
are fully acknowledged, while a check is placed on the abuses pro- 
ceeding from individualism. With the rapid increase in the number 
of Muslims after the conquest of Mecca and battle of Hunayn the 
sense of community between them must have decreased. Muham- 
mad himself, doubtless for political and strategic reasons above 
all, wanted to be on good terms with his former opponents from 
Quraysh. Thus old ties of kinship came to have an increased 
importance within the ummah. 1 Later, as is well known, the deep 
hostility to one another of certain Arab tribes was a major factor 
in the downfall of the Umayyad dynasty. 

In the structure of Islamic society both individualism and com- 
munalism have thus a part. The kinship groups remain important 
for social and administrative purposes, but in the religious sphere 
membership of the clan or tribe has been replaced by membership 
of the Islamic community, and 'tribal humanism* 2 by the religion 
of Islam. 

1 Cf. Q. 8. 75/76; 33. 6; but note Bell's dating. 

2 Cf. Ml Mecca, 24 f., and further Bichr Fares, UHonneur chez les Arabes 
avant V Islam, Paris, 1932. 



IN the course of examining Muhammad's statesmanship and his 
political and social reforms it is easy to forget that he was first 
and foremost a religious leader. A study of his life would be 
incomplete without some account of the religion which he founded. 
Unfortunately the early history of the religious institutions of Islam 
is an obscure and difficult subject. It is therefore most suitable 
here to leave aside the details, and to describe only the general 
features of the institutions. 

The Hijrah brought Muhammad into closer contact with the 
Jews, and thereafter his relations to the Jews determined in large 
measure the line of development taken by the Islamic religion. First 
there was a period of assimilation to Judaism, then a period of 
opposition. 1 To begin with, the thought was that, if the Jews 
fasted, so must the Muslims ; but later it came to be that, if the 
Jews fasted in a certain way, the Muslims must fast in another 
way. Thus we find both ^mil^mesjto Judaism and dissimilarities; 
and in a sense both are deliberate. Indeed this bipolar attitude 
towards the older monotheistic religions, though most apparent 
in the Medinan period, had been present from the earliest days. 
Both religions had political implications which were distasteful 
to Muhammad. His aim was therefore to produce a religion parallel 
to these religions, but specially for the Arabs. The effect of the 
Jewish refusal to recognize Muhammad's religion as in some way 
parallel to their own was that it came to be, not merely a religion 
specially for the Arabs, but also one that was distinctively Arabian. 
The name of Muhammad's religion was not always Islam. In 
the Meccan period one name for it seems to have been tazakki, 
Righteousness', 2 but the religion and its adherents are seldom 
explicitly mentioned. After the Hijrah there are many references to 
'believers' (mu'minUri), 'those who believe', and so forth; in some 
cases these terms include the Jews. On Muhammad's break with 
the Jews he claimed to be following the religion of Abraham, the 

1 Cf. pp. 198 ff. above. * M/Mecca, 165-9. 

304 THE NEW RELIGION * ix. i 

hanif ; and for some time Muhammad's religion must have been 
known as the Hanif lyah. This word was read instead of 'Islam* by 
Ibn Mas'ud in Qur'an, 3. 19/17, x and was presumably the original 
reading. It also occurs in sayings of Muhammad to the effect that 
the religion he took to Medina was the Hanif iyah. 2 This name must 
have had a wide currency, for a Christian writing in Egypt in the 
thirteenth century A.D. can still speak of the time when 'the Hani- 
fite nation appeared and humbled the Romans'. 3 It is difficult 
to say at what period hanif and Hamflyah were replaced by 
muslim and Islam. Richard Bell says that the latter do not occur 
before A.H. 2, 4 and even after their occurrence they may not have 
immediately replaced the others. The variant in the codex of 
Ibn Mas'ud, too, is a reminder that early Medinan passages of the 
Qur'an may have been revised to bring them into line with the 
later nomenclature. 'Islam' is undoubtedly the better designation, 
with a profounder religious content, probably meaning 'resignation 
or submission to God'. It has been suggested that the usage has 
beeif "developed from the account of Abraham's sacrifice of his 
son in the Qur'an, where the two are said to 'resign themselves' 
(aslama). 5 If this is so, then there would be an easy transition from 
'the religion of Abraham' to 'Islam'. 6 

Of the main institutions of Islam, the so-called 'five pillars', the 
most important is the saldt, the Worship or formal Prayer. The 
usual translation of saldt is 'prayers', but this corresponds rather 
to du'd. 7 The Worship had been a feature of Muhammad's religion 
from the earliest times, and attempts to stop his followers from 
worshipping were the first open signs of opposition. 8 The Worship 
did not consist in asking God for favours, b^it was essentially an 
acknowledgement of His might and majesty. 9 ; It was adoration, and 

1 A. Jeffery, Materials for the History of the Text of the Qur'an, Leiden, 

1937, 32- 

2 IH, 411 foot; IS, i/i. 128. 13. Cf. WW, 161 foot, 91 foot ( = IH, 995. n). 

3 Abu alih, The Churches and Monasteries of Egypt, ed. and tr. B. T. A. 
Evetts, 230 f. (quoted from L. E. Browne, The Eclipse of Christianity in Asia, 
Cambridge, 1933, p. 40). 

4 Introduction to the Qur'an, Edinburgh, 1953, 108; contrast J. Horovitz, 
Koranische Untersuchungen, 54 f., following the older dating. 

5 37 103. Cf. H. Ringgren, 'Islam, 'aslama and muslim', Horae Soderblomi- 
anae, ii (Upsala, 1949), 27, &c. ; J. Robson, * "Islam" as a term', Muslim World, 
xliv (1954), 101-9. 

6 Q. 9. 29 I has 'the religion of truth* (din al-haqq). 

7 Cf. E. E. Calverley, Worship in Islam, Madras, 1925, 3. 

8 Q. 96. 10. 9 Cf. Snouck Hurgronje, Verspreide Geschriften, i. 213- 


the fundamental expression of this adoration was a cycle of physical 
acts, repeated twice or oftener, and culminating in the prostration 
(sujud) where the worshipper touched the ground with his fore- 
head. There were forms of words to be repeated along with the 
physical acts at_the_ prostration, for example, something like 
'Praise be to God* but the physical acts were primary and the 
words secondary. This whole conception of worship is very strange 
to the Western European, but in its emphasis on acknowledging 
God the Almighty it is in full harmony with Muhammad's early 
prophetic proclamation. 

When the Worship was stabilized by the later jurists, it became 
obligatory for every Muslim to perform it five times daily. It is 
doubtful, however, whether the five daily hours were regularly 
observed even during Muhammad's closing years. The night- vigil, 
popular with his followers at Mecca, was abrogated at Medina: 1 and 
a phrase in the Qur'an shows that there must have been at least three 
hours of prayer daily. 2 Beyond that nothing certain can be said. The 
call to the Worship was given by the human voice, and the Abys- 
sinian freedman, Bilal, was Muhammad's first muezzin (mu'adh- 
dhiri). The Worship was performed facing in a certain direction, 
theqiblah; at first, as explained above, the worshippers probably 
faced Jerusalem, but after the break with the Jews they turned to 
Mecca. 3 The Worship was preceded by ablutions, 4 and the timing 
for the physical acts was given by a leader (imam) who stood in 
front of the ranks. Performing the physical acts of adoration along 
with one's fellow believers must have fostered a strong sense of 

The Worship might be performed anywhere. It was not neces- 
sary to go to a special place of worship. For the mid-day Worship 
on Fridays, however, there was a strong recommendation that the 
Muslims should gather together in some public place. 5 In Medina 
it became usual to hold this Worship in the courtyard of Muham- 
mad's house; and at other times of the day and week also there 
were probably always a number of the Companions who came to 
join in the Worship with Muhammad. It may be noted in passing 
that, while strangers may sometimes have lodged in the portico 
(suffah, zillah) in this courtyard, the later accounts of a company 

1 Q. 73. 20 and commentaries. 

2 Cf. p. 199 above; also Caetani, Ann. i. 452 f. 

3 Cf. pp. 198, 202 above. 4 Q. 4. 43/46. 5 Cf. Q. 62. 9. 

8783 X 

3<tf THE NEW RELIGION ix. i 

of poor and pious Muslims, known as Ahl as-Suffah, who lived 
there permanently, are not historical. 1 In constructing his residence 
Muhammad probably tried to make it suitable for gatherings of 
the Muslims, religious or secular; but during his lifetime it was 
almost certainly not regarded as a sacred place or sanctuary, like 
the courtyard of the Ka'bah at Mecca. 2 For special occasions Mu- 
hammad went to the musalld or 'chapel' (?) in the district of B. 
Salimah; and it may be that this is the 'place of worship (masjid) 
founded upon piety* to which the Qur'an refers. 3 When Muham- 
mad had died, however, and had been buried in 'A'ishah's apart- 
ment in his residence, it was natural for the residence to become 
the mosque of Medina. Muhammad is also said to have founded 
a mosque among B. e Amr b. 'Awf at Quba', 4 while the story of the 
'mosque of dissension* has already been told. 5 

The second of the five pillars of Islam is the zakat or 'legal alms'. 
The Qur'an frequently employs the phrase 'performing the Wor- 
ship and paying the zakat 9 , and thus indicates that these were the 
distinguishing marks of a good Muslim. Much has already been 
said about the financial aspect of the zakat and its place in Muham- 
mad's budget. It remains to say something of its religious character, 
though this also has been touched on in a discussion of the word 
tazakka. 6 There must be some connexion of thought between 
tazakkd and zakat , but it is difficult to say precisely what it is. 
Zakat, however, was not simply the paying of a tak, though it may 
have appeared in this light to the more secular-minded. It always 
had a religious significance.^ took up into itself old Semitic ideas 
of sacrifice, and provided deep feelings with a form of expression. 
Because of this, zakat has remained one of the pillars of Islam even 
where it ceased to be part of the financial arrangements of the 
state; and in Muslim countries alms-giving is regarded as a pious 

The third of the pillars of Islam is the fast (sawrri) of the month 
of Ramadan. As already related, 7 the practice of fasting was adopted 

1 Cf. art. 'Ahl al-$uffah' in El (2). 

2 IH, 333-7; Tab. i. 1259 f.; Caetani, i. 376-80, 432-47; Buhl, Muhammed, 
204 f. 

3 9. 108/109; A. J. Wensinck, art. 'Mualla' in El (i); C. H. Becker, 'Zur 
Geschichte des islamischen Kultus', Der Islam, iii (1912), 374-99 ( = Islam- 
studien, i. 472-500). 

4 IH, 335; J. Pedersen, art. 'Masdjid', I (a), in El (S). 

5 Cf. p. 190 above. 6 M/ Mecca, 165-9. 
7 Cf. p. 199 above. 


during the Medinan period so that the Muslims might be like the 
Jews, but when the break with the Jews came the Jewish fast of the 
'Ashura was replaced by the fast of Ramadan. 

The fourth pillar is the pilgrimage (hajj), sometimes called the 
'greater pilgrimage' to distinguish it from the 'umrah or 'lesser 
pilgrimage'. 1 In pre-Islamic times the 'umrah seems to have been 
connected specially with the Ka'bah, while the hajj was rather 
associated with other sacred sites in the neighbourhood of Mecca. 
Under Islam the hajj came to be more closely related to the Ka'bah, 
while the 'umrah may be said to consist now of certain supereroga- 
tory works which may be added to the performance of the hajj. 
The verse of the Qur'an enjoining pilgrimage (2. i96/i92a) is dated 
by Bell before Badr; and this adoption of an Arabian custom into 
Islam would suitably occur about the time of Muhammad's break 
with the Jews. Nevertheless something of the kind had long been 
implicit in Islam. An early Meccan revelation (106. 3) recognized 
the Ka'bah as a 'house' of God; and some at least of the Muslims 
who pledged themselves to Muhammad at al- f Aqabah regarded 
the pilgrimage to Mecca as a religious act, and not simply as a con- 
venient excuse for visiting Muhammad. 2 Sa'd b. Mu'adh is said to 
have made the pilgrimage in the first pilgrimage-month after the 
Hijrah, and to have been the guest of Umayyah b. Khalaf. 3 Before 
the next occasion, however, the battle of Badr had occurred, and 
access to Mecca was presumably impossible for all Muslims. 

Nevertheless Muhammad continued to be interested in the 
pilgrimage. The ostensible purpose of the expedition to al-Huday- 
biyah in 628/6 was the performance of the 'umrah, and there is no 
reason to doubt a genuine desire on Muhammad's part to perform 
this religious act, even if he also had political aims in mind. He 
performed the 'umrah a year later according to the treaty, as already 
related, and again in the following year after the battle of Hunayn. 
In 631/9 there was held the first truly Muslim pilgrimage; Abu 
Bakr was the leader, and idolaters were prohibited from taking 
part. 4 A year later in March 632 (xii/io) Muhammad himself 

1 A. J. Wensinck, art. 'tfadjdj' in El (i); R. Paret, art. "Umra', ibid.; C. 
Snouck Hurgronje, Het mekkaansche Feest, Leiden, 1880 (in Verspreide Ge- 
schnften, i. 1-124); M. Gaudefroy-Demombynes, Le Ptlerinage d la Mekke, 
Paris, 1923. 

2 Cf. refusal of al-Bara* b. Ma'riir to turn his back on the Ka'bah, IH, 294 
and p. 169 above. 3 Al-Bukhari, Maghdzi (64), 2; cf. Caetani, i. 425 f. 

4 IH, 919-29; WW, 416 f. Snouck Hurgronje, op. cit., i. 45, suggests that 


led the pilgrimage, 'the pilgrimage of farewell', and established 
the course and form of the ceremonies in general outline; many 
details of what he did, however, were later disputed between the 
various schools of jurists. 

The pilgrimage may be said to focus on one point in space and 
time the whole Islamic world's acknowledgement of the might and 
majesty of God. The recognition of the Ka'bah as the house of 
God par excellence in Arabia (though without any denial of the 
sanctity of Jerusalem) and the adoption of it as qiblah meant that, 
so far as Arabia was concerned, the worship of God was focused 
on one point in space. By retaining the pilgrimage in Islam, albeit 
in a modified form. Muhammad further focused the worship of 
God in time, since the main events of the pilgrimage occur on 
specified days of the pilgrimage-month (DhU 'l-Hijjah). The expan- 
sion of Islam, and the consequent impossibility for the majority of 
Muslims of ever making the pilgrimage, have not led to its becom- 
ing any less central in the Muslim's religious year. The departure 
and return of the pilgrims are great events in many Muslim towns. 
Those who remain at home nevertheless may and do participate 
in one of the ceremonies, the 'festival of the sacrifice' ('id al-adha) ; 
in this they follow the example of Muhammad who, during the 
period when he was debarred from Mecca, annually celebrated 
this feast at the 'place of worship' (musalla) of B, Salimah. 1 The 
climax of the pilgrimage is the 'standing on 'Arafat', a hill and 
plain some four hours' journey east of Mecca by camel. This takes 
place on the Qth of the month from midday to sunset. In recent 
years there have sometimes been half a million people present 
here. The pilgrimage thus makes an important contribution to 
the awareness of Islam as a community, a powerful band of 

The remaining one of the five pillars of Islam is the confession 
of faith (shahadah),that is, the repetition of the words, 'There is no 
god but God, and Muhammad is the Messenger of God'. The 
precise formula does not occur in the Qur'an, though the sense is 
omnipresent. The two points were those on which latterly Muham- 
mad insisted in his dealings with would-be followers, and the 
formula was doubtless used in his closing years. Repetition of it 
made a man a Muslim. 

this measure led by economic pressure to the conversion of the pagans en 
masse. * Tab. i. 1362; Caetani, i. 525; El (i), art. 'Mu^alla', &c. 


In concluding this review of the religious institutions of Islam ' 
attention may be drawn to the individualism which pervades 
them. 1 Worship for the Muslim is essentially something which 
concerns God and the individual only. Where several Muslims are 
together it is appropriate that they, for example, perform the 
Worship together; but when a man is by himself and performs 
the Worship by himself, what he does is just as much a fulfilment 
of God's requirements as what he did in a crowd. In short, all the 
strict obligations of Islamic worship could be carried out even if 
there was only one Muslim in the world. This is a corollary of the 
belief that it is as an individual that man is judged by God on the 
Last Day. Nevertheless, as we have seen, there is also a comple- 
mentary tendency to emphasize the unity of the Islamic com- 
munity, and to develop ceremonies which impress this on the 
worshipper. Perhaps the sum of the matter is that Islam is a com- 
munity of individuals or band of brothers, joined together by 
common duties, but in the last resort not necessary to one 


It is interesting to look more closely at the attitude of Muham- 
mad and the Qur'an to the existing paganism of Arabia, since that 
attitude was more complex than might at first sight appear. As has 
been maintained elsewhere, the vital religious force in the lives of 
most of the Arabs was 'tribal humanism', 2 and the old paganism 
was almost dead. All that remained of it was some magical 
practices and some ceremonies whose meaning had been for- 

In Muhammad at Mecca it was argued that the earliest parts of 
the Qur'an did not contain any attack on paganism, but rather 
assumed in the audience a 'vague monotheism*. ^Later, however, 
the unity of God was strictly insisted on, and a critique of increas- 
ing severity was directed against idolatry.^ This has remained a 
feature of Islam ever since. During Muhammad's Medinan period 
it led to the destruction of idols both in the Medinan clans as some 
members became converted, and in all the chief sanctuaries of the 

1 G. H. Bousquet, Les grandes Pratiques rituelles de V Islam, Paris, 1949, 
116-20. 2 MjMecca, 24 f. 

3 Cf. further C. Brockeimann, 'Allah und die Gotzen, der Ursprung des 
is\amschenMonotheismus\ArchivfurReligionswissenshqft, xxi (1922), 99-121. 

4 P. 63 f., 1 58. 

3 io THE NEW RELIGION ix. a 

Hijaz. Sometimes special raiding parties were sent out; sometimes, 
as in the case of at-Ta'if, the destruction of an idol by its wor- 
shippers was made a condition of their acceptance within the 
Islamic community. 1 In various other ways, too, all vestiges of 
idolatry were removed. Pagan theophoric names were changed; 
thus 'Abd ar-Rahman b. c Awf had originally been 'Abd 'Amr or 
'Abd al-Ka'bah; 2 another 'Abd ar-Rahman had been 'Abd al- 
'Uzza; 3 and so on with many an 'Abdallah. 4 Perhaps the compara- 
tive disuse of the name 'ar-Rahman', 'the Merciful', was due to 
the danger of it encouraging idolatry. For a time in the later part 
of the Meccan period it was being used more frequently than 
'Allah' in references to the Deity (possibly because increased 
emphasis was being laid on the Divine mercy and goodness) ; 5 then 
it passed almost entirely out of use, apart from the heading of the 
surahs. The use of 'ar-Rahman' as a name and not merely an 
epithet could easily cause confusion for simple-minded people, 
especially in view of the association of this name with certain 
localities, such as South Arabia and the Yamamah. 6 

Despite this extirpation of idolatry, many old pagan ideas and 
practices were retained. Though the possibility of the corruption 
of the sources cannot be excluded, it is probable that Muhammad 
himself believed in omens from names. He derived omens of 
success, for example, from the names of the strongholds at Khay- 
bar; 7 and in many instances he is said to have changed inauspicious 
names to the opposite, 'disobedient' to 'obedient', and so on. 8 The 
Qur'an implies belief in the efficacy of cursing, though chiefly, it 
must be admitted, the curse of God. 9 Oaths also were regarded as 
transactions which created a special relation between man and God. 
Even where an oath was part of a practice forbidden by Islam, it 

1 Cf. pp. 69, 103 above. 

2 IS, iii/i. 87 f.; Usd, iii. 313. 

3 IS, iii/2. 41. 18. 

4 e.g. 'Abdallah Dhu '1-Bijadayn, originally 'Abd al-'Uzza (Usd, iii. 122). 

5 H. Grimme, Muhammed, ii. 39 f.; R. Bell, Introduction, 143. 

6 For the latter cf. p. 135 above; Q. 17. no presupposes some confusion. 

7 WW, 272; cf. 266, choice of route. 

8 'Asiyah to Mutf'ah ('disobedient* to 'obedient'), IS, viii. 257. 2; B. as- 
amma' to B. as-Sami'ah ('deaf to 'hearing'), as-Samhudi, i. 138; Khurba to 
alihah, &c. ('waste' to 'prosperous', of a ddr), ibid. 142; al-'Abir to al-Yusayrah, 
Barrah to Juwayriyah, Ju'al to 'Amr, WW, 152, 178, 193. 

9 2. 89/83, &c. ; for human curses cf. 7. 38/36, 2. 155/154; cf. also use of a 
curse in a legal document, IS, i/2. 22. 2 ( 19), 'the curse of God and the angels 
and the people altogether'. 


had to be expiated when broken; 1 and at al-IJudaybiyah the vow 
Muhammad had made to sacrifice the horse of Abu Jahl is said 
to have made it impossible for him to exchange the horse for a 
hundred camels. 2 

The Qur'an also makes it clear that the Muslims continued to 
believe in supernatural beings below the rank of divinity, namely, 
angels, jinn, and demons (shayatm). 3 The precise nature of the jinn 
and their relation to the demons is obscure, but need not be further 
discussed here. On the rejection of the pagan deities, the first 
suggestion was that they were among these lesser supernatural 
beings, but later they were said to be mere names. 4 The forbidding 
of recourse to soothsayers (kuhhari) was doubtless bound up with 
their inspiration by jinn. 5 

The old Semitic idea that certain places were sacred was pre- 
served, at least in a modified form. The sacredness of the Ka'bah, 
apparently recognized in the early revelation (106. 3) where God 
is called 'the Lord of this House', was confirmed by the taking of 
it as qiblah or direction in worship and by the acknowledgement 
of the duty of pilgrimage. The sacredness of the 'sacred mosque* 
(al-masjid al-hardm), however, is based by the Qur'an (22. 26/27) 
on a revelation of God to Abraham, telling him to purify it for 
worship; likewise it is God who makes the neighbourhood of 
Mecca a 'sacred area* (haram). 6 On the contrary, there is nothing 
sacred about the sanctuaries of the pagan deities, and, when the 
sacred objects are destroyed by Muslims and the sites desecrated, 
nothing bad happens to those who do this; thus the nothingness of 
the alleged deities is evident. We see then that although the idea 
has been retained that certain places are sacred, there has been 
a subtle change in it. Places are not intrinsically sacred, and they 
are not made sacred by any of the alleged pagan deities; they only 

1 Q. 58. 3/4, according to the usual interpretation. 

2 WW, 258. 

3 Cf. I. Goldziher, Abhandlungen zur arabischen Philologie, Leiden, 1896, i. 
106-17, &c.; W. Eickmann, Die Angelologie und Ddmonologie des Korans, New 
York, Leipzig, 1908; P. A. Eichler, Die Dschinn, Teufel und Engel im Koran, 
Leipzig, 1928; S. Zwemer, 'The Worship of Adam by Angels', Moslem World, 
xxvii (1937), 115-27. For the connected idea of 'spirit*, cf. D. B. Macdonald, 
'The Development of the Idea of Spirit in Islam', Acta Orientalia, ix (1931), 
6-15 ; T. O'Shaughnessy, The Development of the Meaning of Spirit in the Koran, 
Rome, 1953. 

4 Cf. M/Mecca, 107 f.; Q. 53. 23. 
* WW, 348. 

6 Q. 28. 57; 29. 67; for the Ka'bah cf. 8. 98, &c. 


become sacred through the act of God. In 39 of the Constitution 
there is apparently an extension of this principle, for the valley of 
Yathrib (or Medina) is declared sacred (hararri) by the Messenger 
of God and the Islamic community. 1 

The question of food-restrictions belongs in part to the domain 
of Islam's relationship with paganism. There are several injunc- 
tions to the Muslims to eat what is good and not to regard as for- 
bidden what God had made allowable. 2 Some commentators said 
that this was directed against ascetic practices. 3 The context in 
which several of the passages occur, however, makes it clear that 
they were directed in the first place against pagan taboos. 4 The 
existence of the Jewish food laws led to difficulties, in view of the 
claim that Muhammad's revelation was identical with the Jewish; 
but the difficulties were met partly by insisting that most of the 
Jewish regulations had been instituted as a punishment for them, 5 
and partly by introducing a modified set of restrictions, namely, 
prohibition of what was found dead, of blood, of pork, and of 
animals sacrificed to idols. 6 This list, especially when strangling 
is added, 7 is reminiscent of that adopted by the Christians at 
Jerusalem in the early days of the church meats offered to idols, 
blood, and things strangled 8 and suggests that in this matter 
what was common to Christianity and Judaism was regarded as 
authentic revelation. 

There is no convincing evidence that any belief in magical prac- 
tices was retained in the Qur'an or by Muhammad himself. Islam 
certainly retained rites that had been magical in origin, but the 
Qur'an does not show any signs of belief in their magical efficacy. 
Many Muslims, of course, continued to believe in magic, and 
stories of magical practices have found their way into the traditions 
about Muhammad. One such is about the battle of Badr. Muham- 
mad is said to have taken a handful of pebbles, to have turned 
towards Quraysh expressing a wish for evil on them, and then 
to have thrown the pebbles at them; 9 at his word his followers 
attacked fiercely and Quraysh were routed. This story would not 

Contrast the attitude to the 'sacred month*, p. 299 above. 

Q. 2. 168/163 E; 5. 86/89 f. E; cf. 6. 118 f. B~. 

Cf. Wherry (Sale) on 5. 86/89. 

6. 141/142 ff. c E-f- ; 116 E; ? 148/149 ff. E. 

6. 146/147 E.-f ; 16. n8/ii9f. 

2. 172/167 f. ?F;6. 145/146 E-f ; 16. ii4/nsfF. 7 As in 5.3/4 ? H. 

Acts, xv. 20, 29. 

Nqfaba-hum bi-ha> IH, 445 ; cf. WW, 58. 


deserve much credence apart from the fact that the Qur'an is said 
to refer to the incident in the words : 

Ye did not kill them, but God killed them, and thou threwest not, 
when thou threwest, but God threw. 1 

This passage, however, need not refer to the story of the pebbles; 
the word for 'throw' here (ramaytd) is different from that in the 
story (nafaha) and could easily be applied to shooting with arrows. 2 
Moreover, many of the commentators whose views are recorded 
by at-Tabari appear to be ignorant of the story. 3 This verse, there- 
fore, is no proof of a belief in magic. 

If an attempt is now made to say on what principles pagan ideas 
and practices are retained or rejected, the following conclusions 
may be suggested. Pagan ideas, such as belief in angels, jinn, and 
demons, are retained where they are deep-rooted and do not 
obviously contradict God's oneness. In the case of the idea that 
certain places are sacred, there is also great social utility in the idea 
and in the practices dependent on it the opportunities that it 
provided for the Arabs to meet together in peace, with the resultant 
feeling that they were a single community. These two points of 
deep-rootedness and social utility appear to account for most of 
the beliefs and practices retained. Where neither was present, old 
practices were rejected, such as those connected with camels in 
5. 103/102. 

It is further interesting to note that these pagan survivals, when 
incorporated in Islam, were nearly always transformed. Whatever 
may have been the pagan justification for belief in the sacredness 
of places, the Qur'an made it dependent on God's appointment. 
The angels became servants of God's purposes. The effectiveness 
of curses was probably held to be due to God's activity. The 
lapidation of stone pillars (a rite included in the pilgrimage) was 
interpreted as stoning the devils. Old ideas of sacrifice, as retained 
in the practice of zakdt or 'legal alms', were directed into socially 
useful channels, namely, the relief of the poor or the financing of 
the Islamic state. The actual sacrifice of victims as retained in Islam 
(during the pilgrimage and on one or two other occasions) does not 
have the ideas of atonement and propitiation associated with it as 
in Judaism and Christianity; it is always socially useful, however, 

' Q. 8. 17. 

2 Cf. Lane, s.v.\ also WW, 116, where Muhammad himself used a bow. 

3 Tafsir, ix. 127 f. 


for in theory the sacrificial animal is consumed. 1 There are slight 
traces of these ideas attached to alms-giving, 2 but the later emphasis 
is rather on the reward given by God in the life to come. 3 

Oaths and vows are in a curious position. Although, as already 
seen, they are real transactions and, if broken, have to be expiated, 
yet the final effect of Islam was to render them negligible. An oath 
is essentially the removal of oneself from the protection of one's 
patron deity or exposure of oneself to punishment by this deity. 
But God as conceived by the Muslims cannot be influenced in this 
way by a man's words and deeds. If it becomes clear to a man that 
something he has sworn to do is contrary to God's will or command, 
then it is right to break his oath; the breaking of the oath does not 
separate him from God (or make him liable to punishment) to the 
same extent as disobedience towards God. 4 

On the whole Islam has regarded the outward expression as the 
more important aspect. Where an idea such as belief in the gods 
of paganism had to be renounced, an outward act committing a 
man to the opposite course was demanded or at least encouraged. 
Idols had to be destroyed, both those of the clans and families and 
those in the great sanctuaries. New converts had to take part 
publicly in Islamic worship. On one occasion the leader of a tribe 
which on religious grounds did not eat the heart of animals was 
told that he would only be recognized as chief by Muhammad and 
the Muslims if he ate a heart in their presence. 5 Muhammad's own 
marriage with Zaynab, the divorced wife of his adopted son Zayd, 
was perhaps in part an attempt to demonstrate that such a marriage, 
though contrary to pre-Islamic taboos, led to no evil consequences. 

At the same time, where a pagan practice had nothing obviously 
idolatrous in its outward form, it was comparatively easy for Islam 
to retain it. The typical act of isldm or 'resigning oneself to God' 
was that of Abraham preparing to sacrifice his son in obedience to 
God though the act had no obvious utility. 6 Consequently, in 
accordance with this conception, it was possible for Islam to take 
over practices from pre-Islamic times with little modification, since 

1 Cf. Bousquet, op. cit. H4f.; a Muslim friend, however, tells me that in 
practice nowadays most of the meat of the pilgrimage sacrifices is not consumed. 

2 2. 271/273, 196/192; cf. M/Mecca, 168. 

3 Cf. Q. 2. 276/277; 30. 39/38; &c. 

4 Cf. J. Pedersen, Der Eidbei den Semiten t Strassburg, 1914, 194 ff., esp. 196; 
also Q. 2. 225 ; but cf. p. 328 below. 

5 IS, i/2. 62, &c. 6 Esp. 37. 103, aslamd. 


in order to become Islamic they did not require to have any 
meaning attached to them but only to be regarded as com- 
mands of God. 


The attitude of Islam to the two earlier monotheistic religions 
of the area was closely linked with practical questions. Rela- 
tions with Jews and the Jewish religion have been sufficiently 
studied in Chapter VI. Something has also been said in Chapter IV 
about Muhammad's dealings with Christian tribes, but the ques- 
tion of the Islamic attitude to Christianity, especially as it appears 
in the Qur'an, is worthy of further consideration. 

It is interesting to ask why Muhammad did not become a 
Christian, for the attempt to answer this question brings to light 
important points that are liable to be overlooked. It might seem 
that if, as I have maintained, Muhammad was concerned with the 
social and moral malaise of his time, and looked on the cause of 
this malaise as fundamentally religious, the simplest thing for him 
to do would have been to become a Christian or a Jew. Why did 
he not do this ? The first part of the answer is that in one sense he 
did adopt the religion of Jews and Christians. He originally re- 
garded the monotheism which he believed and preached as identical 
with the existing Jewish and Christian monotheism. 1 The creed 
of the Meccan prophet was not new in itself, but only in respect 
of its adoption and practice at Mecca. This conception of his 
relation to Judaism and Christianity was possible because direct 
contacts were few. There were a few Christians in Mecca, of whom 
one, Khadljah's cousin, Waraqah b. Nawfal, may have influenced 
Muhammad considerably; but the majority were probably Abys- 
sinian slaves and not well instructed in the faith. 2 Muhammad 
would also have seen something of Christianity while trading in 
Syria. Until he went to Medina he may have had practically no 
contacts with Jews. In his first months at Medina he still hoped 
that the Medinan Jews would recognize his prophethood. His 

1 The account of the birth of Jesus in 19. 16-33/34 is in accordance with this 

2 Cf. H. Lammens, 'Les Chretiens a la Mecque a la veille de 1'Ktegire' in 
L* Arabic occidentale a la veille de VHtgire, Beyrouth, 1928, 1-49; C. A. Nallino, 
Raccolta di Scritti, Rome, 1941, iii. 87-156, 'Ebrei e Cristiani nell* Arabia 
preislamica' ; R. Devreesse, 'Le Christianisme dans la Province d 'Arabic', 
Vivre et Fewer, 2 e s6rie, Paris, 1942, 110-46. 


approaches to Christian Arab tribes before the Hijrah 1 may have 
been based on a similar hope for recognition from them. 

Thus for Muhammad the question * Should I become a Chris- 
tian?' did not arise. If during his Meccan period, after he came 
forward publicly as a prophet, the matter had been raised, he would 
have said, 'I acknowledge Jesus the Messiah as a prophet, but my 
business is to preach a similar message to the people of Mecca and 
Arabia.' Before Muhammad received any revelations the question 
might have arisen in the form, 'Should I go to some Christian 
teacher to learn more about God ?' There is no reason for supposing 
that Muhammad did not try to learn as much as possible from 
conversation with Christians such as Waraqah; but a prolonged 
visit to a seminary in Syria or to an outstanding bishop was impos- 
sible. Apart from any financial difficulties, such a visit would have 
had political implications; at the very least the person who acted in 
such a way would have become politically suspect to his fellows. 2 
For the Arab of the Hijaz Christianity was above all the religion 
of the Abyssinians and the Byzantines. To ask formally for instruc- 
tion and baptism would have been to open a channel to these 
foreign influences. It is significant that the opposition to Islam 
about the time of Muhammad's death had as its religious focus 
several Arab prophets who were apparently independent of all 
foreign hierarchies. 3 

Once Muhammad had received a number of revelations, it was 
impossible tor him to accept in full the teaching of Christians and 
Jews without denying the truth of his own revelations. 4 Neverthe- 
less he remained for long in friendly relations with Christians. The 
Negus of Abyssinia gave help and protection to the Muslims who 
'emigrated' to his country. This friendship is reflected in the 
Qur'an(s. 82/85 ff.p): 

Assuredly thou wilt find that the most violent of the people in 
enmity against the believers are the Jews and the idolaters, and thou 
wilt find the nearest in love to the believers to be those who say, 'We are 
Nasara (Christians)'; that is because there are amongst them priests 
and monks, and because they count not themselves great. When they 
hear what has been sent down (revealed) to the messenger (sc. Muham- 
mad), one sees their eyes overflowing with tears because of the truth 
which they recognize; they say, *O our Lord, we believe, so write us 

1 Cf. MfMecca, 140 f. 2 Cf. ibid. 28. 

3 Cf. p. 148 above. 4 Cf. Q. 2. 120/114; &c. 


down among those who bear witness. Why should we not believe in 
God and the truth which has come to us, and crave that our Lord should 
cause us to enter with the upright folk?' So for what they have said, God 
has rewarded them with Gardens through which rivers flow, therein 
to abide; that is the recompense of those who do well. 

While most of the Jews refused to acknowledge Muhammad's 
prophethood, a mixed group of Christians is said to have accepted 
it shortly after Khaybar, and to have been the occasion for the 
revelation of 28. 52-54 ; x while Christians may well have accepted 
Muhammad at that period, it is more likely that the verse is earlier 
and refers to some of the Jewish converts to Islam. 2 

The usual view expressed in the Qur'an in the first few years 
after the Hijrah is that Christianity is a distinct religion parallel 
to Judaism and Islam. 3 The growth of hostility, however, between 
Muslims and Jews did not involve a deterioration of relations 
between Muslims and Christians. On the contrary, we find that 
stories of Jesus the Messiah ('Isa '1-Masih) are used in the Qur'an 
as part of the intellectual attack on the Jews. He is represented 
as having gone to the children of Israel with a message from 
God confirmed by * evidences' (bayyindf) and having been 
rejected by many of them. 4 It is significant that the twelve 
apostles are called the ansdr or 'helpers' of Jesus 5 the name 
applied to the Arabs of Medina who supported Muhammad and 
opposed the Jews; speakers of Arabic probably felt a connexion 
between this word and the Qur'anic word for 'Christians', Nasara. 
Moreover a careful reading of the passage about the crucifixion of 
Jesus shows that it is not intended as a denial of Christian doctrine, 
but as a denial of a Jewish claim to have triumphed over the 
Christians, and it goes on to assert the superiority of the Christian 
hope. 6 

And for their (the Jews') saying, 'We killed the Messiah, Jesus the 
son of Mary, the messenger of God', though they did not kill him and 
did not crucify him, but he was counterfeited for them; verily those 
who have gone different ways in regard to him are in doubt about him ; 
they have no (revealed) knowledge of him and only follow opinion ; and 
certainly they did not kill him, but God raised him to Himself; God 

1 Al-Bay4awi, ad. loc.\ not in at-Tabari. 

2 Tab., Tafsir, xx. 51 f. mentions one Rifa'ah al-Qurazi (perhaps b. SimwSl, 
cf. Usd, ii. 181), and also 'Abdallah b. Sallam. 

3 5- 46/50 f. EF; cf. 5. 69/73. 4 43. 63-65 B; 61.6 B; 5. 110/109-111 F. 
5 3- 52/45-54/47 *-. 6 4- 157/156 f. F- ; cf. 3. 55/48. 


is sublime, wise. There is not one of the People of the Book but will 
surely believe in him before his death, and on the day of resurrection 
he will be a witness against (or regarding) them. 

In the light of this favourable attitude to the Christians at a time 
when the Muslims were hostile to the Jews, we must conclude that 
many of the apparently early Medinan passages criticizing Jews and 
Christians were originally directed only against the Jews. 1 In 2. 
135/129-141/135, for example, the reference to the Christians 
could be removed by the omission of a few words, and there is 
therefore a strong presumption that the passage was later 'revised' 
to make it apply to Christians as well as Jews. There are, of course, 
criticisms in the Qur'an of doctrines specifically held by Christians 
(at least according to Arab ideas). The main point made in the 
Qur'an is that Jesus and his mother are not gods, 2 since Jesus is in 
fact a created being. 3 The view rejected is that Jesus is 'a god* or 
'God'; in the passages referred to there is no mention of Jesus as 
'son of God'. The latter idea is also criticized in the Qur'an, but it 
is given a minor place. The counter-assertion is made that God 
does not beget offspring. 4 This point was first developed as a 
criticism of the term 'daughters of God' applied to the pagan 
deities; and in passages denying that God has offspring the pre- 
sumption is that the primary reference is to paganism unless there 
is a clear mention of Jesus. 5 

The dating of these passages in criticism of the Christians is 
uncertain. Part of the difficulty is that we do not know how far the 
Muslims were acquainted with Christian beliefs prior to the con- 
quest of Mecca. They may not have realized that views resembling 
those mentioned were held by Christians ; or they may have thought 
that such views were the aberrations of a minority, and that the 
great body of Christians regarded Jesus merely as a prophet. In a 
sense, indeed, it is true that the doctrines refuted by the Qur'an, 
namely those of tritheism and of the physical sonship of Jesus, are 
aberrations and not Christian orthodoxy. Thus even if the passages 
criticizing these Christian views are early, they cannot be taken as 
evidence of a generally hostile attitude towards Christians. On the 
contrary, the presumption is that Muhammad maintained friendly 

1 2. 111/105, 120/114; 3. 67/60; 5. 18/21, 51/56. 

2 5. 116-20; cf. 17/19. 3 3. 59/52 ff.; 43. 59- 

4 19- 35/36; 4. 171/169. 

5 e.g. 10. 68/69; 37- 149 ff.J 39- 3/5 ff.J 43. 16/15 ff., 81 ff. 


relations with the Negus of Abyssinia, at least until the return of 
the Muslims at the time of the expedition to Khaybar. 

After the conquest of Mecca and battle of Hunayn the situation 
changed. Muhammad began to have dealings with political groups 
which were wholly or mainly Christian. A letter from him to a 
a certain bishop Dughatir has been preserved, 1 and it looks like 
an attempt to state the tenets of Islam in such a way as to gain the 
bishop's support. It must soon have become clear, however, that 
some Christians, while ready to submit to Muhammad's political 
demands, would never acknowledge him as a prophet to be followed. 
The criticisms of supposed Christian doctrines would be most 
relevant at this period. Political considerations, however, must 
have dominated Muhammad's attitude to Christians. In southern 
and central Arabia the Christian tribes and clans made treaties 
with Muhammad, and at least the weaker among them paid tribute 
to Medina. Along the route to Syria, however, there had been 
a resurgence of Byzantine influence and Muhammad gained hardly 
any adherents here, although expansion into Syria and 'Iraq was 
a strategic necessity for him. The general attitude of Islam to 
Christians came to be determined largely by the attitude to these 
northern tribes, who were nearly all Christian, and friendship was 
replaced by hostility. A revelation came commanding war on them 
until they submitted. 2 

Fight against those who do not . . . practice the religion of truth, of 
those who have been given the Book, until they pay the tribute (jizyah) 
off-hand, being subdued. The Jews say that 'Uzayr (Ezra) is the son of 
God, and the Christians say that the Messiah is the son of God; that 
is what they say with their mouths, conforming to what was formerly 
said by those who disbelieved; God fight them! How they are involved 
in lies! They take their scholars and their monks as Lords apart from 
God, as well as the Messiah, son of Mary, though they were only com- 
manded to serve one God, besides Whom there is no god, glory be to 
Him above whatever they associate (with Him)! They would fain 
extinguish the light of God with their mouths, but God refuses to do 
otherwise than perfect His light, though the unbelievers are averse. 

While parts of this passage were probably revealed on several 
different occasions, the passage as a whole marks the transition to a 
policy of hostility to the Christians. This policy found its expression 

1 IS, i/2. 28 ( 43); translated on p. 358 below. 

2 9- 29-35- 


in the great expedition to Tabuk in 63O/9, 1 and was continued 
not merely for the rest of Muhammad's lifetime but also afterwards, 
at least until Syria had been completely subjugated. In so far as the 
passage prescribes hostility to the Byzantine empire and to Chris- 
tians in general, it long continued to influence the Muslim attitude 
to the Christian church. 

One of the remarkable features of the relationship between 
Muslims and Christians is that neither Muhammad nor any of the 
Companions seems to have been aware of some of the fundamental 
Christian doctrines. Apart from the reference to the crucifixion 
(which is primarily a denial of a Jewish claim), and the mention 
of the twelve apostles as the 'helpers* of Jesus, and of miracles of 
healing and raising the dead, there is nothing in the Qur'an about 
the adult life and teaching of Jesus as recorded in the New Testa- 
ment. The early Muslims gave Jesus the title of Messiah (Masth) 
but did not appreciate that it involved a claim to be ' God's anointed'. 
They did not understand the distinctive work of Jesus in redeem- 
ing the world and atoning for its sins. They did not realize that the 
Holy Spirit was regarded by Christians as the third person in the 
Godhead. It is indeed remarkable that there should have been 
among the Muslims over such a wide area this absence of know- 
ledge of Christianity. The blame for this state of affairs probably 
rests on those Christians with whom Muhammad and his Com- 
panions were in contact, who may themselves have had little 
appreciation of the doctrines mentioned. Nevertheless the 'absence 
of knowledge* remains, and in the thirteen centuries since Muham- 
mad's time few Muslims have done anything to fill in the lacuna. 

1 Cf. p. 116 above. 




SEVERAL accounts have been preserved of the appearance of 
Muhammad, and, as they largely agree, they are perhaps near 
the truth, though there is a tendency in some of them to paint 
a picture of the ideal man. 1 According to these accounts Muham- 
mad was of average height or a little above the average. His chest and 
shoulders were broad, and altogether he was of a sturdy build. His 
arms, or perhaps rather forearms, were long, and his hands and 
feet rough. His forehead was large and prominent, and he had 
a hooked nose and large black eyes with a touch of brown. The hair 
of his head was long and thick, straight or slightly curled. His 
beard also was thick, and he had a thin line of fine hair on his 
neck and chest. His cheeks were spare, his mouth large, and he 
had a pleasant smile. In complexion he was fair. He always walked 
as if he were rushing downhill, and others had difficulty in keeping 
up with him. When he turned in any direction, he did so with his 
whole body. 

He was given to sadness, and there were long periods of silence 
when he was deep in thought; yet he never rested but was always 
busy with something. He never spoke unnecessarily. What he said 
was always to the point and sufficient to make his meaning clear, 
but there was no padding. From first to last he spoke rapidly. Over 
his feelings he had a firm control. 2 When he was annoyed he would 
turn aside; when he was pleased, he lowered his eyes. His time was 
carefully apportioned according to the various demands on him. 
In his dealings with people he was above all tactful. He could be 
severe at times, but in the main he was not rough but gentle. His 
laugh was mostly a smile. 3 

1 IS, i/2, 120-31 ; cf. WW, 349 f. 

2 But cf. WW, 373 f., where he strikes a man, perhaps because he was over- 
strained, and later gives him a present. 

3 The accounts of his aversion to poetry and inability to scan it may contain 
some truth but are suspect because of their 'tendency' to enhance the miraculous 
character of the Qur'fin. Cf. Q. 36. 69 ; IH, 882 ; WW. 376 ; Ibn Hanbal, Musnad, 
i. 134, 148, 189; &c. He certainly disliked hostile poets, but he encouraged 
favourable ones like Hassan b. Thabit, Ka'b b. Malik, 'Abdullah b. Rawahah 
(IS, ml i. 80 f.). 

6788 Y 


There are many stories illustrating his gentleness and tenderness 
of feeling. Even if some of them are not true, the probability is that 
the general picture is sound. There seems to be no reason, for 
instance, for doubting the truth of the story of how he broke the 
news of the death of Ja'far b. Abi Talib to his widow Asma' bint 
c Umays; the story is said to have been told by Asma' herself to her 
grand-daughter. 1 She had been busy one morning with her house- 
hold duties, which had included tanning forty hides and kneading 
dough, when Muhammad called. She collected her children she 
had three sons by Ja'far washed their faces and anointed them. 
When Muhammad entered, he asked for the sons of Ja'far. She 
brought them, and Muhammad put his arms round them and smelt 
them (as a mother would a baby). Then his eyes filled with tears 
and he burst out weeping. 'Have you heard something about 
Ja'far?', she asked, and he told her that he had been killed. Later 
he instructed some of his people to prepare food for Ja'far's house- 
hold, 'for they are too busy today to think about themselves'. 
About the same time the little daughter of Zayd b. Harithah (who 
had been killed along with Ja'far) came to him in tears to be com- 
forted, and he wept along with her; afterwards, when questioned 
about this, he said it was because of the great love between Zayd 
and himself. 2 The memory of his first wife Khadljah could also 
soften his heart. After Badr the husband of his daughter Zaynab 
was among the prisoners taken by the Muslims, and Zaynab sent 
a necklace of Khadijah's to Muhammad for a ransom, but he was so 
moved at the sight of it that he set the man free without payment.3 

Muhammad seems to have felt especial tenderness towards 
children, and to have got on well with them. 4 Perhaps it was an 
expression of the yearning of a man who had seen all his sons die 
in infancy. Much of his paternal affection went to his adopted son 
Zayd, who has just been mentioned. He was also attached to his 
nephew 'AH b. Abi Talib, who had been a member of his house- 
hold for a time, but he doubtless realized that 'All had not the 
makings of a successful statesman. Among the stories showing his 
affection for children are some about his grand-daughter, Umamah 
bint Abi 'l-'As (the daughter of Zaynab). He sometimes carried her 
on his shoulder during the Worship, setting her down when he 
bowed or prostrated, then picking her up again. On one occasion 

1 IS, via. 206. z IS, iii/i. 32. 5. 3 WW, 77. 

4 I am indebted to Sir H. A. R. Gibb for calling my attention to this point. 


he teased his wives by showing them a necklace and saying he 
would give it to the one who was dearest to him; when he thought 
their feelings were sufficiently agitated, he presented it not to any 
of them but to Umamah. 1 He was also fond of Zayd's son Usamah 
and took him on his camel for a bit when he returned from the 
battle of Badr. 2 

He was able to enter into the spirit of childish games and had 
many friends among children. f A'ishah was still a child when he 
married her, and she continued to play with her toys. He would 
ask her what they were. 'Solomon's horses', she replied, and 
Muhammad smiled. 3 He is even said to have had a game of 'spit- 
ting* with a child. 4 He had fun with the children who came back 
from Abyssinia and spoke Abyssinian. 5 We hear of a house in 
Medina where there was a small boy with whom he was accus- 
tomed to have jokes, for it is recorded that once he found the small 
boy looking very sad; when he asked what was the matter, he was 
told that his pet nightingale had died, and he did what he could to 
comfort him. 6 In view of all this kindness and liking for children, 
the following story may be true, even though it has a legal import. 
A baby was once brought to Muhammad ; he took it in his arms, and 
in due course it wet him. When the mother slapped it, he re- 
proached her saying 'You have hurt my son', and this is the 
legal point refused to change his clothes to have them washed, 
since this was not necessary in the case of a boy baby. 7 His kindness 
extended even to animals, and this is something remarkable for 
Muhammad's century and part of the world. As his men marched 
towards Mecca just before the conquest they passed a bitch with 
puppies, and Muhammad not merely gave orders that they were 
not to be disturbed, but posted a man to see that the orders were 
carried out. 8 

These are interesting sidelights on the personality of Muham- 
mad, and fill out the picture of him we form from his conduct of 
public affairs. He gained men's respect and confidence by the 
religious basis of his activity and by such qualities as courage, 
resoluteness, impartiality, firmness inclining to severity but tem- 
pered by generosity. In addition to these, however, he had a 

1 IS, viii. 26 f. a WW, 72. 

3 IS, viii. 42. 16; further references in N. Abbott, Aishah, 8. 

4 Usd, v. 393. 6 from foot. 5 IS, iv/i. 72. 
6 IS, iii/2. 65. 12. 7 IS, viii. 204. 

8 WW, 327. 


charm of manner which won their affection and secured their 


Of all the world's great men none has been so much maligned 
as Muhammad. It is easy to see how this has come about. For 
centuries Islam was the great enemy of Christendom, for Christen- 
dom was in direct contact with no other organized states compar- 
able in power to the Muslims. The Byzantine empire, after losing 
its provinces in Syria and Egypt, was being attacked in Asia Minor, 
while Western Europe was threatened through Spain and Sicily. 
Even before the Crusades focused attention on the expulsion of 
the Saracens from the Holy Land, medieval war-propaganda, free 
from the restraints of factuality, was building up a conception of 
'the great enemy*. At one point Muhammad was transformed into 
Mahound, the prince of darkness. By the eleventh century the 
ideas about Islam and Muslims current in the crusading armies 
were such travesties that they had a bad effect on morale. The 
crusaders had been led to expect the worst of their enemies, and, 
when they found many chivalrous knights among them, they were 
filled with distrust for the authorities of their own religion. It was 
to deal with this situation that Peter the Venerable started the 
process of disseminating more accurate information about Muham- 
mad and his religion. Since then much has been achieved, especially 
during the last two centuries or so, but many of the old prejudices 
linger on. 1 

In the modern world, where there are closer contacts than ever 
before between Christians and Muslims, it is urgent that both 
should strive to reach an objective view of Muhammad's character. 
The denigration of him by European writers has too often been 
followed by romantic idealizations of his figure by other Europeans 
and by Muslims. The aim of the present discussion is to work 
towards a more objective attitude with regard to the moral criti- 
cisms inherited from medieval times The main points are three. 
Muhammad has been alleged to be insincere, to be sensual, and to 
be treacherous. 

1 Cf. G. Pfannmliller, Handbuch der Islamliteratur, Berlin, 1923, 133-97; 
Montgomery Watt, 'Carlyle and Muhammad', Hibbert Journal, liii (1954-5), 
247 ff. ; the views of Islam taken by Latin writers from the twelfth to fourteenth 
centuries are studied by N. A. Daniel in an Edinburgh Ph.D. thesis. 


The allegation of insincerity or imposture was vigorously 
attacked by Thomas Carlyle over a hundred years ago, has been 
increasingly opposed by scholarly opinion since then, and yet is 
still sometimes made. The extreme form of the view was that 
Muhammad did not believe in his revelations and did not in any 
sense receive them from 'outside himself, but deliberately com- 
posed them, and then published them in such a way as to deceive 
people into following him, so gaining power to satisfy his ambition 
and his lust. Such a view is incredible. Above all it gives no satisfy- 
ing explanation of Muhammad's readiness to endure hardship in 
his Meccan days, of the respect in which he was held by men of 
high intelligence and upright character, and of his success in 
founding a world religion which has produced men of undoubted 
saintliness. These matters can only be satisfactorily explained and 
understood on the assumption that Muhammad was sincere, that is, 
that he genuinely believed that what we now know as the Qur'an 
was not the product of his own mind, but came to him from God 
and was true. 

This conception of Muhammad's sincerity, however, is open to 
possible misunderstandings and requires to be made more precise. 
Thus, to say that Muhammad was sincere does not imply accep- 
tance of the Qur'an as a genuine revelation from God; a man may 
without contradiction hold that Muhammad truly believed that he 
was receiving revelations from God but that he was mistaken in 
this belief. Further, once this point is grasped, it should be clear 
that, even if true, the alleged fact that the revelations fitted in with 
Muhammad's desires and pandered to his selfish pleasure would 
not prove him insincere ; it would merely show him to be capable 
of self-deception. The verses usually quoted in this connexion are 
33. 37 f., justifying his marriage with Zaynab bint Jahsh, and 
33. 50/49, granting him special marriage privileges. The affair of 
Zaynab will be considered presently, and it will be shown that 
Muhammad was not merely yielding to selfish desires In connexion 
with the other verse 'A'ishah is said to have made the remark, 'God 
is in a hurry to satisfy your desires'. 1 Even if she really said this 
(and it is not a later invention), it would only show that she was 
suspicious of the correspondence between the revelation and Mu- 
hammad's desires; but, as the remark itself suggests that she was 

1 IS, viii. 112. 5, 141. 2; Ibn Hanbal, vi. 134, 158; al-Bukhari, Tafstr (65) 
on Q. 33. 50/49. 


jealous, she cannot be taken as an impartial witness On the con- 
trary, if a remark like this could be made to Muhammad without 
disturbing his belief in himself, that tends to confirm the view that 
he was sincere. 

Again, the theory held by Richard Bell and others, that Muham- 
mad 'revised' passages of the Qur'an, is not necessarily at variance 
with a belief in his sincerity. The revision, if it may be so called, 
consists in the addition or omission of words, phrases, and longer 
passages. Muhammad may be presumed to have regarded these 
changes as emendations communicated to him by God to meet 
fresh circumstances. A certain amount of revision is admitted by 
Muslim orthodoxy in its doctrine that some verses have been 
abrogated, that is, have ceased to be applicable to the Muslims. 
Additions could be justified in a similar way; for example, God 
could reveal to Muhammad that the words 'and Christians' were 
to be added to a verse about Jews, and the justification for the 
change would be that, whereas the Muslims had at first to deal 
only with Jews, latterly they had to deal with Christians also ; the 
words 'and Christians' might simply have been confusing if in- 
cluded in the original revelation at a time when they had no 
practical application. Muhammad had possibly some technique of 
'listening' or 'waiting* for an emending revelation, and he may 
have employed this also when there was a topic on which he felt 
that a revelation was desirable. Whatever his technique, however, 
when words 'came to him' he had some means of knowing when 
they were from God. To say he was sincere is simply to hold that, 
when he had thus recognized words as being from God, he really 
believed that they were so, and did not confuse them with his own 

There are no sufficient grounds, then, for regarding Muhammad 
as an impostor. On the contrary, the case for his sincerity is strong. 
A high degree of certainty is attainable here, since the discussion, 
unlike that of the other two moral criticisms, is at a factual level 
and does not involve any dispute about moral standards. In other 
words, if it could be shown that Muhammad, in full knowledge 
that the contents of the Qur'an were the product of his own mind, 
gave out that they were revelations to him from God, that would 
be imposture and would be generally recognized as a serious moral 

When we come to the other two allegations, however, namely, 


that Muhammad was morally defective in that he was treacherous 
and sensual, the discussion has to embrace not merely factual 
points, but also the question of the standard by which the acts have 
to be judged. On the factual side, there is agreement on such acts 
as his breaking of the treaty of al-Hudaybiyah and his marriage to 
Zaynab, the divorced wife of his adopted son, but there is ample 
room for dispute about circumstances and motives. With regard 
to standards there are two main possibilities : we may ask, * Was 
Muhammad a good man according to the standards of the Arabia 
of his day ?', or we may ask, 'Was he a good man according to the 
standards of, say, the best people in Europe about the year 1950?' 
Let us begin, then, by trying to answer the first of these questions 
with special reference to the two points of criticism. 

The allegation of treachery may be taken to cover a number of 
criticisms made by European writers. It applies most clearly to 
such acts as the breaking of his agreements with the Jews and his 
one-sided denunciation of the treaty of al-Hudaybiyah with the 
Meccans. 1 It may also, however, be taken to include the infringe- 
ment either of the sacred month or of the sacred territory on the 
expedition to Nakhlah when the first Meccan blood was shed, the 
mass execution of the Jewish clan of Qurayzah, and the orders or 
encouragement given to his followers to remove dangerou oppo- 
nents by assassination. 2 

In all these actions there was nothing which disturbed the con- 
science of Muhammad's followers apart from the events at Nakhlah. 
This may seem incredible to the European, but that is in itself 
a measure of the remoteness of the moral ideals of ancient Arabia 
from our own. In some respects the nomadic Arabs had a high 
ideal of conduct, but they had no idea whatsoever of a minimum 
standard of decent behaviour towards all men, simply because 
they were men. They had no conception of a universal moral law 
of the Kantian type. There were customary duties and obligations 
within the tribe (and this included those attached to the tribe as 
confederates, clients, or slaves); related to these matters was an 
ideal of honourable conduct. Outside the tribe, however, there 
were no duties or obligations. You could do what you liked with 
an unprotected stranger. When you were at war with another tribe, 
it was a case of * nothing barred*. The only restraints on your 

1 Cf. pp. 208 ff., 63 above. 

2 Cf. pp. 8, 214, 15 above. 


behaviour towards an enemy or even a stranger were those set by 
fear of retaliation or fear of supernatural powers. 

Now the Islamic community or ummah was thought of as a tribe. 
Towards tribes with which it had agreements, it had duties and 
obligations, and these were scrupulously observed according to the 
standards of the day ; Muhammad even paid blood-money to a man 
who was really but not technically responsible for the death of 
several Muslims. 1 Where a tribe was at war with the Muslims, 
however, or had no agreement, they had no obligations towards 
it even of what we would call common decency. If contemporaries 
showed some surprise at the execution of all the males of Qurayzah, 
it was because Muhammad was not afraid of any consequences of 
such an act; the behaviour of Qurayzah during the siege of Medina 
was regarded as having cancelled their agreement with Muham- 
mad. Similarly, the terms of the treaty of al-Hudaybiyah had been 
broken by the Meccans before Muhammad denounced it, and the 
individuals who were assassinated had forfeited any claim to 
friendly treatment by Muhammad through their propaganda 
against him. So far were the Muslims who killed them from feeling 
any qualms that one of them, describing the return from the deed, 
wrote that they returned with the head of their victim 'five honour- 
able men, steady and true, and God was the sixth with us'. 2 This 
is so much in keeping with the spirit of pre-Islamic times that it 
is almost certainly authentic; but, even if not, it shows the attitude 
of the early Muslims. 

One point at which the actions of Muhammad or the Muslims 
were disapproved for no merely selfish reasons was the expedition 
to Nakhlah. The disapproval, however, was not on moral but on 
religious grounds; something sacred had been violated, and those 
who disapproved were presumably afraid of supernatural punish- 
ment. In this instance Muhammad deliberately decided in the end 
to oppose public opinion, believing that he was thereby following 
God's command and attacking a pagan uperstition. In the cir- 
cumstances, then, Muhammad's conduct is that of a moral reformer 
and not of a wicked man. Similarly, in respect of oaths, the religious 
aspect was probably more important than the purely moral one. 
The Qur'an (66. 2) permits the annulling of oaths, and for reasons 
already suggested oaths came to have little importance in Islam. 3 

1 Cf. p. 32 above. * WK, 190; WW, 971 cf. Q. 58. 7/8. 

3 Cf. p. 314 and n. 4 above. 


In so far as the keeping of one's word in pre-Islamic times had been 
associated with pagan deities, and since oaths by God did not fully 
take the place of pagan oaths, there may have been a vacuum in 
which there were no effective sanctions for keeping one's word. 

Again, the common European and Christian criticism that Mu- 
hammad was a sensualist or, in the blunter language of the seven- 
teenth century, an 'old lecher', fades away when examined in the 
light of the standards of Muhammad's time. There was a strain in 
early Muslim thought which tended to magnify the common or 
perhaps we should say 'superhuman* humanity of their prophet. 
There is even a tradition to the effect that his virility was such that 
he was able to satisfy all his wives in a single night. 1 This looks like 
an invention, for the usual account is that he gave his wives a night 
each in turn, but it shows the outlook of some at least of his 
followers. The early Muslims looked askance at celibacy and 
checked any movements towards it, and even rigorous ascetics in 
Islam have commonly been married. Compared with the late king 
'Abd al-'Aziz of Arabia, known as Ibn Sa'ud, Muhammad was 
temperate in the matter of wives. His contemporaries thought none 
the less of him for the multiplicity of his marital relations ; to them 
it would be no more than what was befitting a man of his political 
power. These contemporaries or their immediate successors even 
seem to have touched up the stories about his relations with 
women, such as those about the jealousies in his hareem and his 
'love at first sight' for Zaynab. 

The one point of this kind on which Muhammad was criticized 
by his contemporaries was his marriage with Zaynab bint Jahsh. 
Zaynab was Muhammad's cousin, being the daughter of his 
father's sister. At the time of the Hijrah she was either unmarried 
or (more probably) a widow, and she went to Medina, presumably 
with her brothers. Muhammad made her, against her will, marry 
his adopted son, Zayd b. Harithah. Some time afterwards, about 
the year 626/4 Muhammad called at Zayd's house to talk to him; 
Zayd was out, but he saw Zaynab in disarray, and is supposed to 
have been smitten by love for her. He went away saying to himself, 
'Praise be to God, praise to the Manager of Hearts!' Zaynab told 
Zayd about Muhammad's visit, his refusal to enter, and his cryptic 
utterance. Zayd at once went to Muhammad and offered to divorce 
Zaynab, but Muhammad told him to keep his wife. After this, 

1 IS, i/z. 96. 


however, life with Zaynab became unbearable for Zayd, and he 
divorced her. After her waiting-period ('iddah) had been observed, 
a marriage with Muhammad was arranged, and justified by a 
revelation. 1 

About the main outline of the story there can be little dispute, 
but the significance of the various actions is a matter for discussion. 
One point is tolerably certain, and that is the reason for the criti- 
cism of Muhammad's action by his contemporaries. They were not 
moved in the slightest by what some Europeans have regarded as 
the sensual and voluptuous character of his behaviour. They were 
opposed to the marriage because in their eyes it was incestuous. 
This view of the marriage was doubtless based on the Qur'an, 2 in 
conjunction with the old principle that an adopted son counted 
as a real son. We cannot be certain of all that is involved, but the 
most natural explanation of the Qur'anic passages is to suppose 
that there was something objectionable about the equating of 
adoptive sons with real sons, and that it was desirable that there 
should be a complete break with the past in this respect. 3 The 
Qur'an implies that Muhammad had originally been unwilling to 
marry Zaynab and afraid of public opinion, but had come to 
acknowledge the marriage as a duty imposed on him by God; his 
marriage demonstrated to the believers that there was no blame 
in marrying the divorced wife of an adoptive son. 4 The criticism 
of Muhammad, then, was based on a pre-Islamic idea that was 
rejected by Islam, and one aim of Muhammad in contracting the 
marriage was to break the hold of the old idea over men's conduct. 
How important was this aim compared with others which he might 
have had ? 

It is not too much to say that all Muhammad's marriages had a 
political aspect. 5 There is therefore a strong presumption that in the 
case of Zaynab bint Jahsh Muhammad was not carried away by 
passion but was looking at the political implications of the match. 

1 IH, 1002; IS, viii. 71, 81; Q. 33. 37; Caetani, i. 6iof.; Abbott, Aishah, 

2 4. 23/27; cf. p. 280 above. Cf. also Wellhausen, 'Die Ehe bei den Arabern', 
441, n. 3. 

3 33. 4 f. S. Kohn, Die Eheschlieflung im Koran, 12, notes that if Muhammad 
had merely wanted to marry Zaynab, he could have made this a khdli$ah, special 
privilege, for himself; since he made it a general rule, other points must be 
involved. Cf. G. H. Bousquet, in Studio, Islamica, ii. 78. 

4 33- 37- 

5 Cf. p. 287 above. 


There are two points of importance : Zaynab was a close relative of 
Muhammad's, and her family were, or had been, confederates of 
Abu Sufyan's father. As Zaynab's marriage took place long before 
that with Abu Sufyan's daughter and at a time when Abu Sufyan 
was directing the Meccan campaign against Muhammad, this aspect 
of the match cannot have escaped Muhammad. It is also clear, how- 
ever, from the records that he used the marriages of his cousins, 
like those of his daughters, for political ends. Just as Fatimah was 
married to ' All, and Ruqayyah and later Umm Kulthum to 'Uth- 
man b. 'Affan, so Hamnah bint Jahsh was married after Uhud to 
Talhah b. ' Ubaydallah and Habibah bint Jahsh to 'Abd ar- Rahman 
b. 'Awf. There can therefore be no doubt that the marriage of 
Zaynab to Zayd was part of this scheme of alliances, since Zayd 
was a prominent man in the community, in some ways as prominent 
as Abu Bakr. 

Beyond this point doubts increase. Why was Zaynab unwilling 
to marry Zayd ? She can hardly have thought that he was not good 
enough. She was an ambitious woman, however, and may already 
have hoped to marry Muhammad; or she may have wanted to 
marry someone with whom Muhammad did not want his family 
to be so closely allied. After the incident of Muhammad's visit to 
Zayd's house, Zaynab clearly worked for marriage with Muham- 
mad. What of Muhammad's reasons for marrying her at this 
particular time ? It cannot be that Zayd was declining in his favour, 
because in 627/6 and subsequent years Zayd led several expeditions, 
including the large one to Mu'tah on which he met his death. 
Perhaps he realized that she was tired of Zayd, and had no follower 
worthy of becoming her husband. Perhaps he felt that the time 
had come when he was strong enough to go against public opinion 
and contract this marriage that was politically and socially desir- 
able. Despite the stories, then, it is unlikely that he was swept off 
his feet by the physical attractiveness of Zaynab. The other wives 
are said to have feared her beauty ; but her age when she married 
Muhammad was thirty-five, or perhaps rather thirty-eight, which 
is fairly advanced for an Arab woman. 1 The only one of his wives 
who was older at marriage was Khadijah. 

In general, then, there was nothing in Muhammad's marital 

1 IS, viii. 8 1 f.; she is said to have been 35 when she married Muljammad in 
A.H. 5, but she is also said to have been c thirty odd' at the Hijrah and 53 when 
she died in A.H. 20. 


relationships which his contemporaries regarded as incompatible 
with his prophethood. They did not consider him a voluptuary 
any more than they considered him a scoundrel. The sources record 
criticisms of him, but these are based on no moral criterion, but 
on a conservatism which was akin to superstition. Though later 
Muslims might produce colourful stories of Muhammad's sus- 
ceptibility to feminine charm, and though there is no reason to 
suppose that he disregarded the factor of physical attraction, it is 
practically certain that he had his feelings towards the fair sex well 
under control, and that he did not enter into marriages except when 
they were politically and socially desirable. 

It is possible, too, to go further and, while restricting oneself to 
the standpoint of Muhammad's time, to turn the alleged instances 
of treachery and sensuality into matter for praise. In his day and 
generation Muhammad was a social reformer, indeed a reformer 
even in the sphere of morals. He created a new system of social 
security and a new family structure, both of which were a vast 
improvement on what went before. In this way he adapted for 
settled communities all that was best in the morality of the nomad, 
and established a religious and social framework for the life of 
a sixth of the human race today. That is not the work of a traitor 
or a lecher. 

It may be remarked at this point that there solid grounds 
for thinking that Muhammad's character declined after the Hijrah. 
Too facile a use has been made of the principle that all power 
corrupts and that absolute power corrupts absolutely. The allega- 
tions of moral defects are attached to incidents belonging to the 
Medinan and not the Meccan period ; yet, if the exposition just 
given of these incidents is sound, they marked no failure in Mu- 
hammad to live up to his ideals but were in accordance with his 
moral principles. As the ruler of Medina was a man of his time, so 
also was the persecuted preacher of Mecca. If nothing is recorded 
of the preacher to show us how different his attitude was from that 
of nineteenth-century Europe, it does not follow that his ideals 
were any loftier than those of the reforming ruler. The opposite is 
more likely to be the case in so far as the preacher was nearer to 
the pagan background. In both Meccan and Medinan periods 
Muhammad's contemporaries looked on him as a good and upright 
man, and in the eyes of history he is a moral and social reformer. 

Up to this point Muhammad has been considered only in relation 


to the moral standards of his time, but there is also another way 
of judging him, namely, by a universal moral standard. I do not 
propose here to attempt any judgement of this sort, but shall be 
content if I have fairly presented the evidence on which such 
a judgement must be based. The readers of this book will presum- 
ably include Christians and Muslims of many different shades of 
opinion, as well as persons who are neither, and, even if there is 
a wide area of agreement between them, there are also differences 
which make it impossible in a book like this adequately to meet all 
the objections that any such judgement would arouse. There is one 
thing, however, which may be said in this connexion. 

The world is becoming increasingly one world, and in this one 
world there is a tendency towards unification and uniformity. 
Because of this tendency the day will doubtless come when there 
will be a set of moral principles which not merely claim universal 
validity but are actually accepted almost universally throughout 
the one world. Now Muslims claim that Muhammad is a model of 
conduct and character for all mankind. In so doing they invite 
world opinion to pass judgement upon him. Up till now the matter 
has received scant attention from world opinion, but, because of 
the strength of Islam, it will eventually have to be given serious 
consideration. Are any principles to be learnt from the life and 
teaching of Muhammad that will contribute to the one morality 
of the future ? 

To this question the world has not yet given a final answer. What 
has been said so far by Muslims in support of their claims for 
Muhammad can be regarded as no more than a preliminary state- 
ment of the case, and few non-Mutlims have been convinced by it. 
Nevertheless the issue still remains open. How the world answers 
the question about Muhammad depends to some extent on what 
the Muslims of today do. They still have an opportunity to give a 
fuller and better presentation of their case to the rest of the world. 
Will they be able to turn to the life of Muhammad and by sifting 
the universal in it from the particular discover moral principles 
which make a creative contribution to the present world situation ? 
Or, if this is too much to expect, will they at least be able to show 
that Muhammad's life is one possible exemplification of the ideal 
man in the unified world morality? If they make a good case, 
there are some Christians who will be ready to listen to them and 
to learn whatever is to be learned. 


The difficulties confronting Muslims, however, are immense. 
A combination of sound scholarship and deep moral insight is 
essential, and this combination is rare. I will not conceal my 
personal view that Muslims are unlikely to be successful in their 
attempt to influence world opinion, at least in the sphere of morals. 
In the wider sphere of religion they have probably something to 
contribute to the world, for they have retained emphases on the 
reality of God, for example which have been neglected or for- 
gotten in important sections of the other monotheistic religions ; 
and I for one gladly acknowledge my indebtedness to the writings 
of a man like al-Ghazali. Towards convincing Christian Europe 
that Muhammad is a moral exemplar, however, little, indeed 
nothing, has so far been accomplished. 


Circumstances of place and time favoured Muhammad. Various 
forces combined to set the stage for his life-work and for the subse- 
quent expansion of Islam. There was the social unrest in Mecca 
and Medina, the movement towards monotheism, the reaction 
against Hellenism in Syria and Egypt, the, decline of the Persian 
and Byzantine empires, and a growing realization by the nomadic 
Arabs of the opportunities for plunder in the settled lands round 
them. Yet these forces, and others like them which might be added, 
would not in themselves account for the rise of the empire known 
as the Umayyad caliphate nor for the development of Islam into 
a world religion. There was nothing inevitable or automatic about 
the spread of the Arabs and the growth of the Islamic community. 
But for a remarkable combination of qualities in Muhammad it is 
improbable that the expansion would have taken place, and these 
vast forces might easily have spent themselves in raids on Syria 
and 'Iraq without any lasting consequences. In particular we may 
distinguish three great gifts Muhammad had, each of which was 
indispensable to the total achievement. 

First there is what may be called his gift as a seer. Through him 
or, on the orthodox Muslim view, through the revelations made 
to him the ^^j^dd wasjgjven, an ideological framework within 
which the resolution of its social tensions became possible. The 
provision of such a framework involved both insight into the 
fundamental causes of the social malaise of the time, and the genius 
to express this insight in a form which would stir the hearer to the 


depths of his being. The European reader may be 'put off* by the 
Qur'an, but it was admirably suited to the needs and conditions 
of the day. 

Secondly, there is Mubammad'sjvisdom as a statesman. The 
conceptual structure found in the Qur'an was merely a framework. 
The framework had to support a building of concrete policies and 
concrete institutions. In the course of this book much has been 
said about Muhammad's far-sighted political strategy and his 
social reforms. His wisdom in these matters is shown by the rapid 
expansion of his small state to a world-empire and by the adaptation 
of hijs social institutions to many different environments and their 
continuance for thirteen centuries. 

Thirdly, there is his skill and tact as an administrator and his 
wisdom in the choice of men to whom to delegate administrative 
details. Sound institutions and a sound policy will not go far if the 
execution of affairs is faulty and fumbling. When Muhammad died, 
the state he had founded was a 'going concern', able to withstand 
the shock of his removal and, once it had recovered from this shock, 
to expand at prodigious speed. 

The more one reflects on the history of Muhammad and of early 
Islam, the more one is amazed at the vastness of his achievement. 
Circumstances presented him with an opportunity such as few 
men have had, but the man was fully matched with the hour. 
Had it not been for his gifts as seer, statesman, and administrator 
and, behind these, his trust in God and firm belief that God had 
sent him, a notable chapter in the history of mankind would have 
remained unwritten. It is my hope that this study of his life may 
contribute to a fresh appraisal and appreciation of one of the 
greatest of the 'sons of Adam'. 

Further Remarks on the Sources 

LITTLE need be added to what was said about the sources in the 
Introduction to Muhammad at Mecca. 1 There are one or two points, 
however, which will bear further emphasis. 

The first point is the relation of the Qur'an to the other source- 
material. I have made use of the Qur'an wherever possible, but the 
small proportion of references to it will make it clear that the 
historian has to rely mainly on other material. Nevertheless, so 
far as it goes, the Qur'an is a first-hand witness, especially to the 
contemporary feelings of the Muslim community about various 
events. It is also the main witness for the reform of the social 
system. The passages revealed at Medina can frequently be dated 
on the basis of internal evidence. The most complete and satis- 
factory attempt to do this is that of Richard Bell in his translation 
of the Qur'an. In several places where the dating was not obvious 
the investigations involved in writing this book have confirmed 
Bell's results (which were presumably reached along different 
lines) ; and I have therefore adopted the position that in general 
his' dating is to be regarded as authoritative. 2 

The sources other than the Qur'an may be termed collectively 
'the traditional historical material'. In Muhammad at Mecca 3 the 
attitude taken towards this material was that in general it is to be 
accepted; only where there is internal contradiction is it to be 
rejected; where 'tendential shaping' is suspected it is as far as 
possible to be corrected. Perhaps the coherence of the resulting 
account of Muhammad's career will be accepted as an additional 
argument for the soundness of this procedure. An important appli- 
cation of the principle adopted is the acceptance as genuine of 
Muhammad's letters and treaties (with the exception of the first 
seven) as reproduced by Ibn Sa'd (i/z). 

While, then, the traditional historical material is to be regarded 
as in general sound, a distinction must be made within it between 
material concerning disputed points and material where there is 
no reference to any disputed point. This distinction may be roughly 

1 pp. xi-xvi. 

2 For the system of indicating it see p. xiii above. 3 p. xiv. 


described as one between legal material and historical material. 
Where an anecdote about Muhammad involved legal issues dis- 
puted between rival schools, it was liable to be twisted a little by 
each to make it support their views. This has happened very often 
in traditions with a legal bearing, but it is also found where the 
point at issue is rather theological or political. Thus in the story 
of Muhammad's call 1 he is stated to have used the words ma aqra'u 
which may mean either 'I cannot read* or 'What am I to recite?' 
The latter is probably the original meaning, but certain later 
theologians insisted on Muhammad's inability to read as a confir- 
mation of the doctrine of the miraculous nature of the Qur'an, and 
versions are found where words have been substituted which can 
only mean 'I cannot read* (viz. ma ana bi-qdriri). The story has 
thus been given a little twist in order to make a theological point. 
At the same time rival theologians have given it a little twist in the 
other direction by replacing ma by mddhd, so that it can only mean 
'What shall I recite?' Again, the remark of Muhammad about 
Sa'd b. Mu' adh when he was about to judge the case of Banu Quray- 
zah, 'Stand for your chief (sayyid), could be taken to justify the 
view that the Ansar were capable of ruling over Quraysh, and 
the story was therefore twisted in various ways to remove this 
implication. 2 

On the other hand, where a story involved no disputed point of 
this sort, the presumption is that it was not twisted, or at least not 
twisted to the same extent. Details of how an ancestor behaved 
during Muhammad's lifetime would often be preserved in a clan 
or family, and naturally they would dwell on what glorified the 
descendants and omit anything of an opposite character. Such 
exaggerations and omissions, however, are much easier to correct 
than the twists given to stories in legal and other disputes. The 
critique of Islamic traditions by European scholars, notably Ignaz 
Goldziher in his Muhammedanische Studien and Joseph Schacht 
in his Origins of Muhammadan Jurisprudence, has been based 
mainly on the legal traditions found in the standard collections by 
al-Bukhari, Muslim and others, that is, on the section of the 
traditional historical material where distortion is most to be 
expected. It is thus not surprising if sceptical views about the 

1 M/Mecca, 46. 

2 Cf. Montgomery Watt, 'The Condemnation of the Jews of Bana Quray?ah' 
Muslim World, xlii, 1952, 164. 

6783 Z 


traditions have resulted. If, however, one considers the undisputed 
or purely historical section of the traditional historical material, it is 
apparent that there is a solid core of fact. 1 

The process of transmission may be conceived somewhat as 
follows. To begin with the stories would be handed down infor- 
mally in families and clans, and from the older men and women to 
younger acquaintances. Before the end of the first Islamic century, 
however, a few persons had begun to collect all the information 
they could about the life and campaigns of Muhammad, and some 
at least wrote down what they had collected. These early collectors 
of information, however, though they seem to have scrutinized 
their sources carefully and sometimes stated what they were, did 
not in every case give a complete isndd or chain of authorities going 
back to an eye-witness of the events. It was only gradually that the 
complete isndd became de rigueur. Ibn Ishaq, working in the second 
quarter of the second Islamic century (middle of the eighth century 
A.D.), usually gives his authorities, but not always a complete chain, 
and he does not always repeat the words of the authority verbatim. 
Al-Waqidi, half a century later, is similar in method, but his 
secretary and follower, Ibn Sa'd, some twenty years younger, 
always attempts to quote exactly and to give a complete chain of 
authorities. The insistence on complete chains is to be associated 
with the teaching of ash-Shafi'i, 2 who was roughly a contemporary 
of al-Waqidi. Once it became fashionable to give complete isnads, 
scholars must have been tempted to extend their chains backwards 
to contemporaries of Muhammad. Even when they thus added to 
the chains, however, their additions may have been sound, since 
they probably knew in a general way where their predecessors had 
obtained information. This means only that we cannot rely so fully 
on the early links of a chain as on the later ones. 3 

1 Cf. Montgomery Watt, op. cit. 171. The distinction was noted by J. 
Horovitz, 'Alter und Ursprung des Isnad', Der Islam, viii (1918), 39. 

2 Cf. Schacht, op. cit. 

3 Cf. Muslim World, I.e. For earlier discussion of the sources cf. J. Flick, 
Muhammad b. Ishaq, Frankfurt, 1925; J. Horovitz, 'The Earliest Biographies of 
the Prophet and their Authors', Islamic Culture, i (1927), 535-59; J. Schacht on 
Musa b. 'Uqbah in Ada Orientalia, xxi (1953), 288-300; A. Guillaume in 
a forthcoming volume on Ibn Isfraq. 


List of Expeditions and Dates 

N o independent chronological investigations have been undertaken 
in connexion with this study of the life of Muhammad, since the 
disputed points hardly affect the general picture. The first main 
point in dispute is whether the Muslims observed intercalary 
months during the first ten years at Medina. Intercalation was 
forbidden at the pilgrimage at the end of the year 10, and therefore 
it may be taken as certain that the first day of the year 1 1 corres- 
ponded to 29 March 632. Without intercalation, then, the begin- 
ning of the era of the Hijrah (that is, the first day of the year in 
which Muhammad migrated to Medina) would be on 16 July 622. 
Even if the Muslims observed intercalary months (presumably 
three) during the ten years, it is almost certain that statements in 
the sources are made on the basis of orthodox Muslim reckoning, 
with no intercalation, since scholars in the second Islamic century 
would overlook intercalation or deliberately reject it. Orthodox 
reckoning also fits in best with a number of statements in the 
sources. 1 

The order and dating of some of the separate expeditions is the 
other main point of dispute. Ibn Ishaq gives a number of dates, 
but the first complete chronology is that in al-Waqidl. The best 
course is that adopted by Leone Caetani, namely, to follow al- 
Waqidi as a general rule where there are discrepancies between 
him and Ibn Ishaq. 2 The Shi'ite leanings of al-Waqidi presumably 
do not affect his chronology. 











or Name 



Number of 












Hyrah, arr 


of era of Hi 
val in Qubl 






no fighting 



1 For a defence of this position see Caetani, Ann. i. 345-60. The position is 
apparently admitted by H. Amir 'All, 'The First Decade in Islam', Muslim 
World, xliv. 136. 

2 The discrepancies are discussed by Caetani, Ann. i. 466, SiQf., 575, 577, 
ii. 509 f., &c. The following list is based on his Annali; cf. his Chronographia 
Islamica, Paris, 1012, i. 

3 Heavier type indicates years and normal type months. The day of the month 












or Name 



Number of 













no fighting 







Sa r d b. Abi 


no contact 

422 f. 

33 f. 



(or 8) 







no contact 

415 f- 









no contact 






Kurz al- 

failed to 











no contact 

4 2Xf. 









one enemy 



b. Jatsh 







c. 315 











995 f. 

90 f. 







994 f- 

91 f. 













200 or 400 


543 f. 










some booty 

539 f. 

94 f. 




Ka'b b. 







mad b. 









? no con- 


99 f- 




tact but 















Zayd b. 



547 f. 

100 f. 



















c. 900 

enemy re- 






<D x 




Aba Sala- 






(?) x 


Sufyftn al- 





22 4 f. 


b. Unays 








































Badr al- 



no contact 






Abu RafT 




9 8x 


b. Unays 

is occasionally placed before a stroke; thus 16/7 means the 1 6th day of the seventh 
month (that is, in Christian dates, July). Where the disagreement between the 
Muslim and Christian months was slight, it has been neglected. 











or Name 



Number of 









Dhat ar- 




no contact 








DQmat al- 



some booty 


174 ^ 













































mad b. 








no contact 

718 f. 

226 f. 






























2 3 2 f. 



mad b. 



















Zayd b. 

. . 


J t 


JamQm 1 








469 f. 

233 f. 







c. 15 

















Wadi '1- 

(Badr b. 








1 2-1 

DQmat al 


'Abd ar- 


Kalb sub- 

236 f. 




b. 'Awf 


1 2-1 







237 f. 





Badr b. 

Zayd b. 



238 f. 








Usayr b. 





239 f. 


b. Rawa 










240 f . 

Fihrl (?) 




. . 

Madyan 8 

Zayd b. 


. . 

























. . 

Aban b. 

. . 

. . 


Cf. Caetani, i. 694 f . 

Ibid. 705. 

3 Ibid. ii. 56. 













or Name 



Number of 












no contact 








AbQ Bakr 








Bashir b. 









Ghalib b. 



297 f. 















Bashir b. 










. . 










Ibn Abi 












Ghalib b. 




307 f. 







? Qu<;la'ah 

Ka'b al- 


all but one 














308 f. 





Zayd b. 









Dhat as- 


'Amr b. 




315 f. 







Sif at-Bahr 

part of 



did not see 


317 f- 














(Jus ham) 






to north 



no contact 



I ^ am 



















Hisham b. 









Khalid b. 








Khalid b. 








(various expeditions from Mecca to destroy shrines) 

839 f- 

350 f. 


























373-8 X 

tion of 












b. Hisn 





Qutbah b. 



. . 







b. Sufyan 









successful ? 


388 f. 














or Name 



Number of 










Tayyi 1 





389 f. 


al-Hubab 1 









indecisive ? 






DGmat al- 


Khalid b. 












AbO Bakr 


416 f. 




Khalid b. 




(417 n.) 


b. Ka'b 










967 f., 

















a success- 



b. Zayd 

ful raid 



death of Muhammad 




1 Cf. Caetani, ii. 235. 


Slaves and Freedmen among the Emigrants at Badr 

1. Zayd b. Harithah (261) :* an Arab, captured when young and 
sold into slavery; freed by Muhammad. 

2. Anasah (33) : freedman of Muhammad, born in slavery, said 
to be of Persian father and Abyssinian mother. 2 

3. Abu Kabshah (33) : freedman of Muhammad, born in slavery, 
presumably Arab. 

4. Salih Shuqran (34) : an Abyssinian slave, bought by Muham- 
mad from 'Abd ar-Rahman b. *Awf and given positions of trust. 
(The notice says that there were other slaves at Badr belonging to 
c Abd ar-Rahman, Hatib b. Abi Balta'ah and, from the Ansar, 
Sa'd b. Mu'kdh.) 

5. Salim (60): freedman and adopted son of Abu Hudhayfah, 
of Persian descent. 

6. Sa'd (81): freedman of Hatib b. Abi Balta'ah, said to be 
Sa'd b. Khawlayy, of the tribe of Kalb or Madhhij, and to have 
been enslaved on capture. 

7. Khabbab b. al-Aratt (i 16) : born free but sold at Mecca after 

8. Suhayb b. Sinan (161) : an Arab but capturfed by Byzantines ; 
either sold to Kalb and by them to 'Abdallah b. Jud'an who freed 
him, or escaped to Mecca and put himself under the protection 
of 'Abdallah b. Jud'an. 

9. 'Amir b. Fuhayrah (164) : born in slavery (according to Usd y 
iii. 90); bought by Abu Bakr and freed. 

10. Bilal b. Rabah (165) : of Abyssinian descent, born in slavery, 
freed by Abu Bakr. 

11. Mihja' b. Salih (285): from the Yemen (presumably Arab); 
fell into captivity; freed by 'Umar b. al-Khattab. 

1 The details are from Ibn Sa'd, iii/i, and the number in brackets after each 
name is a reference to the page there. 2 Tab. i. 1780, 7. 


Muhammad's Letters to the Princes 

THE position has been adopted 1 that the material collected by 
Ibn Sa'd in volume i/2, pp. 15-86 is in general to be regarded as 
authentic. An exception must be made, however, of the story with 
which the collection opens, that in May 628 (i/y) on his return from 
al-Hudaybiyah Muhammad sent six messengers to the rulers of the 
surrounding countries summoning them to accept Islam. The 
messengers were c Amr b. Umayyah ad-Damri, Dihyah b. Khalifah 
al-Kalbl, 'Abdallah b. Hudhafah as-Sahmi, Hatib b. Abl Balta'ah, 
Shuja* b. Wahb al-Asadi, and Salit b. 'Amr al-'Amiri; and they 
were sent respectively to the Najashi or Negus of Abyssinia, the 
governor of Bostra (Busra) as representative of the Byzantine 
emperor, the Persian emperor, the Muqawqis or ruler of Alex- 
andria, a Ghassanid prince called al-Harith b. Abl Shimr, and 
Hawdhah b. 'All of the tribe of Hamfah in the Yamamah. 2 

This story cannot be accepted as it stands. 3 Muhammad was 
a wise and far-seeing statesman, and he did not 'lose his head* after 
the measure of success he obtained at al-Hudaybiyah. To appeal 
to these princes at this period to accept Islam would have done 
more harm than good. Moreover, close examination shows that 
the sending of some of the envoys was prior to al-Hudaybiyah. The 
mission of Dihyah to Bostra must have been in the summer of 627, 
since he was plundered by Judham on his return and a punitive 
expedition was sent against them about October 627 (vi/6).* The 
two slave-girls brought back by the envoy to the Muqawqis appear 
to have been in Medina soon after January 627 (viii/5), since 
Muhammad presented one of them to Hassan b. Thabit at the 
conclusion of 'the affair of the lie'. 5 Further, it is possible to discern 
a theological motive for the alteration of the stories. Ibn Ishaq 
makes Muhammad himself refer to the sending out of the apostles 
by Jesus, and with this connects the gift of languages at Pentecost. 

1 Cf. p. 336 above. 

2 The story of Farwah b. 'Amr must be a later intrusion, since Muhammad 
did not take the initiative in his case; cf. below. J. Wellhausen, Skizzen, iv. 102 
noted the break after the seventh. 

3 Cf. Caetani, i. 725-39; Buhl, Muhammed, 294-8. 

4 Cf. p. 43 above. s Cf. p. 186 above. 


This appears to be intended to substantiate the claim that Muham- 
mad was a prophet to all nations and not simply to the Arabs. 1 

The conclusion to be drawn is not that the stories of the six 
envoys are worthless, but that they contain a kernel of fact which 
has become distorted in the course of transmission because of the 
theological interest. This factual basis is clearly discerned in the 
case of the envoy to Abyssinia ; he had to arrange for the marriage 
of Muhammad to Umm Habibah and perhaps also for the return 
to Arabia of Ja'far b. Abi Talib and the other Meccans who still 
remained there. In the case of the others it is practically certain 
that the aim of the embassy was to conclude a friendly agreement 
between Muhammad and the ruler in question. Such agreements 
would be primarily political, though there would probably also be 
some mention of religion similar to that in the letter to Bishop 
Dughatir. 2 There is difficulty about the identification of the Mu- 
qawqis and of al-Harith b. Abi Shimr, 3 but this is not surprising 
since this was about the time of the Persian withdrawal from Egypt 
and before the Byzantines had regained full control. It is also 
doubtful whether any envoy went to the court of the Persian 
emperor; the kernel of fact in this story is the agreement with the 
Persian governor in the Yemen. 4 

It would be interesting to study in detail all the early material 
for these embassies contained in Ibn Hisham, Jbn Sa'd and at- 
Tabari. These extant sources raise questions about the circles 
responsible for the theological manipulation of the original accounts 
and about the relationship between the recensions of Ibn Ishaq by 
Ibn Hisham and at-Tabari respectively. Here it must suffice to 
indicate the general course of the 'tendential shaping* of the 
material. Ibn Ishaq appears to have made a list of the envoys 
Muhammad 'sent in different directions to the princes (muluk) of 
the Arabs and the non- Arabs, summoning them to God, in the 
period between al-Hudaybiyah and his death 1 . s This list included 
the six named above, and also the names of three sent to 'Uman, 
al-Bahrayn, and the Yemen. 6 Ibn Ishaq was also aware of the 

1 IH, 971 f.; Tab. i. 1560; Buhl, 295, with a further reference to his art. 
'Fafite Muhammed seine Verkiindigung als eine universale, auch fur Nicht- 
araber bestimmte Religion auf ?' in the Festschrift for A. Fischer (Islamica, ii). 

2 Cf. p. 358 below. 

3 Cf. Caetahi, i. 730, 735; Buhl, 296 f., nn., with further references. 

4 Cf. 122 above. 5 Tab. 1560. 5 f. 

6 IH, 971 ; all except the one to the Yemen are mentioned in Tab. 1560-71. 


parallel with Jesus, but does not seem to have emphasized it or 
brought the stories into accordance with it. Al-Waqidi either 
reduced the list to six, or had an independent list. 1 The three 
omitted the envoys to 'Uman, al-Bahrayn, and the Yemen 
were to minor Arabic-speaking princes and were known not to 
have been sent until after Hunayn. Ibn Hisham seems to have 
shortened Ibn Ishaq's narrative, to have corrected one or two slips, 
and to have brought out more clearly the parallel with Jesus. The 
story of Farwah, first included by Ibn Sa'd (or possibly al-Waqidl) 
is probably intended to replace that of the mission of Salit to Haw- 
dhah, since the latter ruled in east-central Arabia (and so was 
hardly a foreign potentate), while Farwah was in the Byzantine 

empire. 2 

1 Tab. 1559. 13-1560. 3; cf. IS, i/2. 15 

2 Ibid. 18. i(6);cf. 31 (53). 


' Those whose hearts are reconciled' 

THE phrase from Q. 9. 60, al-mu j allafah qulubu-hum, 'those whose 
hearts are reconciled', is commonly applied to the leading Meccans 
(along with one or two others), either still pagans or recent converts 
to Islam, who received 50 or 100 camels from Muhammad during 
the distribution of spoils at al-Ji'ranah. This use of the phrase is 
at least as early as Ibn Ishaq, who writes : 'The Messenger of God 
made gifts to the mu'allafah qulftbu-hum, who were some of the 
leaders of the people, reconciling them (yata'allafu-hum) and 
through them reconciling their tribe.' 1 The suggestion is that it 
was only this substantial gift that made these men accept Islam ; 
Safwan b. Umayyah is alleged (probably falsely) to have said, 
'Muhammad was the most hateful of men to me, but he made me 
a gift, and at once he was the dearest of men to me.' 2 

Study of the Qur'anic passage, however, shows that it cannot 
apply to the men of al-Ji'ranah. The normal meaning of mu'allafah 
would be 'are reconciled', that is, 'are already reconciled'; the 
meaning 'are to be reconciled', though perhaps possible, is unusual. 
Further, if the Qur'an referred to persons who were wavering, it 
would not refer to them by any term with a Suggestion of dis- 
paragement; this would have defeated the purpose of winning 
them over. The decisive point, however, is that the Qur'an is deal- 
ing with the sadaqdt or contributions of the Muslims, whereas the 
gifts made to the men of al-Ji'ranah were admittedly made either 
from the fifth of the spoils, which was Muhammad's share, or 
from the fard'i\ what remained over after four camels or their 
equivalent had been given to every man in the army. This is stated 
by al-Waqidi, and al-Bukhari deals with the matter under the 
heading of Khums 3 At-Tabari in his commentary takes the view 
that the phrase refers to the men of al-Ji'ranah, but the traditions 
that he quotes do not all support him, since the first, sixth, seventh, 
and eighth run as follows : 

(i) Ibn 'Abbas said . . . these are some people who used to come to 
Muhammad, having been converted, and he would make a little gift 

1 1H, 880; cf. IS, ii/i. no. 10. 

2 T a b., Tafsir, on the phrase (x. 98-99), third tradition. 

3 WW, 367; al-BukhSri, Khums (57), 19. 


to them (rakhada, which implies something much smaller than 100 
camels) from the sadaqdt ; when he did this and they obtained something 
good from the $adaqdt, they said, This is a right religion ; otherwise they 
would have spoken ill of it and left it. (6) People of the Bedouin and 
others whom Muhammad used to reconcile with gifts so that they 
believed. (7) Jewish and Christian converts . . . even if rich. (8) Jews or 
Christians. 1 

This divergence of view shows that the Qur'anic verse did not refer 
clearly and unambiguously to the gifts made at al-Ji'ranah. The 
first tradition would fit in better with the gifts that Muhammad 
made to the various 'deputations' which came to him from the year 
8 onwards. Whatever the original application of the phrase, it 
cannot have been a justification of Muhammad's conduct towards 
Abu Sufyan and the others at al-Ji'ranah. 

Can any further light be thrown on how it came about that the 
men of al-Ji'ranah are normally described as 'those whose hearts 
are reconciled'? The matter is not clear, but some stages seem 
traceable. In particular 'Uyaynah b. Hisn and al-Aqra' b. Habis, 
two nomadic chiefs, have special prominence. Writing in the 
thirteenth/seventeenth century with marks of judicious scholar- 
ship, al-Baydawi comments thus : 

Al-nttf allafah qulubu-hum. Some people who had become Muslims 
but whose resolve was weak, so (Muhammad) tried to reconcile their 
hearts. Or else : leading men to whom gifts were given and regard shown 
in the expectation of the conversion of their fellows; it was for that 
purpose that Muhammad had made gifts to ' Uyaynah and al-Aqra e and 
al-'Abbas b. Mirdas. Another view is: leading men whom he tried to 
reconcile so that they became Musliirs, for Muhammad used to make 
gifts to such men. But the truth is rather that these gifts were made (not 
from the adaqdt but) from that fifth which was his private property. 

With this may be compared at-Tabari's statement of his own views 
on the phrase : 

These are some people who were reconciled to Islam, but to whom 
it was not right to give anything in order to improve their condition or 
that of their tribe (sc. since they were not needy) ; for example, Abu 
Sufyan, 'Uyaynah, al-Aqra f and their fellows among the chiefs of tribes. 

The special prominence of 'Uyaynah and al-Aqra' (along with 
al-'Abbas b. Mirdas) is to be connected with a story told by both 
Ibn Ishaq and al-Waqidl. 2 'Uyaynah b. Hisn (of Fazarah) had long 

1 Tab., To/sir, ad loc. * IH, 88 1 ; WW, 376. 


been a thorn in the flesh to Muhammad. Both he and al-Aqra* b. 
Habis (of Tamim), who now appears for the first time, accom- 
panied Muhammad on his expedition to Mecca and Hunayn, but 
perhaps as observers rather than as allies, since they did not have 
their tribesmen with them. Nevertheless at al-Ji'ranah Muhammad 
treated them like the chiefs of the Meccan clans and gave them 
100 camels each. Upon this al- f Abbas b. Mirdas of Sulaym, whose 
tribe had sent over 900 men, complained that he had received only 
four camels, though his ancestry was no less noble than that of 
'Uyaynah and al-Aqra'. The point may have been that he had been 
treated as an ordinary man and not as a chief, or (if we suppose 
that 100 camels was the recognized share of the leader of a non- 
Muslim group allied to Muhammad) it may have been that Sulaym 
was to be reckoned as an allied group. The memory of a recent war 
between Ghatafan (to which Fazarah belonged) and Sulaym would 
make al- 'Abbas feel the supposed slight more keenly. 1 He expressed 
his dissatisfaction in verse. When this was reported to Muhammad, 
he gave orders for the docking of his tongue, which was understood 
to mean presenting him with 100 camels. 

This story may well have a basis in fact, though the point does 
not affect the present argument. Al-Bukhari has a tradition which 
seems to refer to this incident, though without naming al-'Abbas. 
'Abdallah (b. Mas'ud), after remarking how thfe Prophet showed 
special favour to al-Aqra' and 'Uyaynah and others (unnamed), 
told how a man came and accused him of being unjust, and how 
Muhammad exclaimed, 'Who is just, if not God and His Messen- 
ger ?' It is noteworthy that one of the important links in the isndd 
is a Sulami, Mansur b. al-Mu'tamir (d. 132), who also had leanings 
towards the Shi'ah. 2 

There is another story in which the same two chiefs are singled 
out for mention. 3 Someone asked Muhammad why he made gifts 
to those two, and did nothing for Ju'ayl (or Ju'al) b. Suraqah (of 
Damrah). He replied that Ju'ayl was much better than fellows like 
these, 'but I treated them well (td } allaftu-huma) that they might 
become Muslims, whereas I trusted Ju'ayl to his faith*. (Ju'ayl was 
given lodging, at least for a time, within the mosque at Medina 
(that is, Muhammad's house), and performed various errands for 

1 Caussin de Perceval, ii. 556-64. 

2 Khums, 19. 8. 

3 IH, 883 ( = Tab. 1681 f.); IS, iv/i. 181, from al-Waqidi. 


him; 1 he may have lived in Medina before he became a Muslim, 
though his tribe was one of the first to make an alliance with 
Muhammad; he was apparently of low morale at the battle of 
Uhud, for he told Muhammad of a dream in which he had seen him 
killed, and later, perhaps because of the dream, he was thought 
to have raised the shout that Muhammad was killed.) Ibn Ishaq 
reports this story on the authority of Muhammad b. Ibrahim b. 
al-Harith (b. 119-21), a Medinan lawyer or jurisconsult, belonging 
to the clan of Taym of Quraysh, and grandson of one of the early 
Muslims who had migrated to Abyssinia. Such a man doubtless 
belonged to the 'pious opposition* in Medina, and the point of the 
story and the selection of Ju'ayl may be to emphasize the superi- 
ority of faith to worldly goods. Some of the other stories about the 
dissatisfaction of the Ansar with the distribution of spoil at al- 
Ji'ranah have the same point. The fact that the same words 'with 
them I treated a group of people well, that they might become 
Muslims, but I trusted you to your faith* are applied to the Ansar 
as a whole in another story, 2 suggests that at least one of the stories 
is an invention. Nevertheless it is probably true that Muhammad's 
gifts to 'Uyaynah and al-Aqra e (as distinct from those to Abu 
Sufyan and Hakim b. Hizam and perhaps others) were intended to 
win them to accept Islam. To this extent Mujahid is justified in 
interpreting the Qur'anic phrase as meaning "people whom he 
reconciled by gifts, 'Uyaynah and those with him'. 3 

A complicating factor is the story of how Muhammad, from 
some gold sent to him by 'All from the Yemen, made presents to 
four chiefs, al-Aqra', 'Uyaynah, Zayd at-Ta'i, and 'Alqamah b. 
'Ulathah al-'Amiri, and, when Quraysh and the Ansar complained, 
said, 'I am reconciling them (ata'allafu-huni).'* With this is joined 
a version of the story of how Muhammad was accused of being 
unjust. The latter is suspect, since there are many forms of it and 
it was developed as propaganda against the sect of Khawarij ; but 
the distribution of the gold may be largely true. 

The next stage in the process we are trying to disentangle is the 
connexion of the two desert chiefs with Abu Sufyan and other 
Quraysh who received 50 or 100 camels at al-Ji'ranah. We have 
already seen how at-Tabari in his summing-up mentioned only 

' WW, 204, 272, 407; cf. IS, I.e. 

2 IH, 886. 8. 3 Tab., Tafsir, ad. loc., fourth tradition. 

4 Al-Bukhari, Anbiyd' (60), 6; cf. Muslim, Zakdt, 143. 


Abu Sufyan along with 'Uyaynah and al-Aqra'. In the Sahlh of 
Muslim there are earlier instances of a similar character. 1 One 
version says that Muhammad gave 100 camels to Abu Sufyan, 
Safwan b. Umayyah, c Uyaynah and al-Aqra c , and fewer to al- 
' Abbas b. Mirdas; other versions omit Safwan or add 'Alqamah. 
The imdd of some is Kufan, with 'Umar b. Sa'id b. Masruq, 
brother of Sufyan ath-Thawri, as principal figure. One version 
has a Medinan imdd which includes 'Amr b. Yahya b. 'Umarah 
(d. 757/140); this version alone has the words 'and he made gifts 
to the mu'allafah qulubu-hum'. 

Finally, there are the (more or less) complete lists. There are two 
in Ibn Hisham. 2 One is on the authority of Ibn Ishaq alone; the 
other came to Ibn Hisham from an unnamed informant, who had 
it from Ibn Shihab az-Zuhri, from 'Ubaydallah b. 'Abdallah b. 
'Utbah, from Ibn 'Abbas. The second of the traditions in aj- 
Tabari's commentary on the phrase, which gives a list of thirteen 
names, is traced back to Ma'mar (b. Rashid) (d. 769-71/152-4) 
and Yahya b. Abi Kathir (d. 746-50/129-32). Al-Waqidi does not 
give any authority for his whole list, but certain details have come 
through az-Zuhri. 3 Thus everything points to the conclusion that 
the identification of 'those reconciled in heart' with the men who 
received gifts at al-Ji c ranah was made in Medina among the 'pious 
opposition' not later than the early years of the second century A.H. 
Ma'mar is noted as one of the best exponents of the views of 
az-Zuhri, and Yahya must have belonged to the same circle. 

It need not be seriously doubted that az-Zuhri and his friends, 
many of whom had been supporters of the revolt of 'Abdallah b. 
az-Zubayr and were still hostile to the Umayyads, should have 
found great delight in spreading stories which put the ancestors 
of the Umayyads and their supporters in an unpleasant light. It 
was almost certainly anti-Umayyad feeling which brought about 
the change of interpretation of the phrase al-mu'allaf ah qulubu-hum. 
While blackening the Umayyads, az-Zuhri did his best to avoid 
blackening his own friends. He could not deny that Hakim b. 
Ilizam, now head of the clan of Asad (the clan of az-Zubayr), had 
received 300 camels, but he continued by telling how, as a result of 
Muhammad's words on this occasion, he came so to despise 
worldly goods that he never again accepted the slightest gift; in the 

1 Zakdt, 137-9. 2 880-1, 882-3. 

3 WW, 375-6. 


case of Makhramah b. Nawfal of his own clan of Zuhrah, he denied 
that he received any gift on this occasion. 1 

The course of development suggested by this study of the 
traditions may be outlined as follows. At an early period two 
stories were in circulation; whether true or not, we need not ask, 
though it seems probable that they were true. One was about the 
complaint made by al- c Abbas b. Mirdas against 'Uyaynah and al- 
Aqra c at al-Ji f ranah, and the other was about a gift of gold to these 
chiefs and some others, presumably not at al-Ji'ranah. As part of this 
second story, or perhaps at first independently of it, Muhammad was 
said to have made a remark using the word ta'allaftu/I reconciled*. 
There was some conflation of the stories. The word tcCallaftu came 
to be applied to 'Uyaynah and al-Aqra* in connexion with al- 
Ji'ranah, and was then extended to all those who received a large 
gift there, though for some it was not suitable and led to lexico- 
graphical difficulties. From this, as mu'allafah comes from the 
same root, it was a short step to the assumption that the Qur'anic 
phrase meant the men of al-Ji'ranah, especially since, presumably, 
it no longer had any practical application. 

In all this it is not denied that Muhammad was well aware of the 
importance of material inducements in attracting men to Islam 
and frequently made use of them. This was probably so in the case 
of many of the recipients at al- Ji'ranah. What is mischievous, how- 
ever, is the suggestion that this was the dominant factor in the 
submission of the leaders of Quraysh. It keeps one from appreciat- 
ing one aspect of Muhammad's achievement, the fact that, leading 
an army which included these Meccans, he had triumphantly 
averted a threat, with apparently overwhelming force, from the 
old enemy of Mecca. 

1 Ibid.; cf. al-Bukhari, Khums, 19. i. 

6783 A a 


Text of Selected Treaties 

THE following are translations of some of the more interesting 
treaties and letters preserved by Ibn Sa'd. 1 All the treaties and 
letters in this volume were edited and translated into German 
by Julius Wellhausen in Skizzen und Vorarbeiten, iv, section 3 
(Berlin, 1889). His interpretation has usually been followed. 

1. The Messenger of God (God bless and preserve him) wrote 
to B. Damrah b. Bakr b. c Abd Manat b. Kinanah that they have a 
guarantee for (dminun 'ala) their goods and persons, and that 
succour (nasr) is due to them against whoever wrongfully oppresses 
them ; the succour of the prophet (God bless and preserve him) is 
(incumbent) on them so long as water wets a piece of wool, unless 
they are (already) fighting about God's religion; 2 when the prophet 
summons them, they are to respond to him. On that condition, 
there is over them the dhimmah 3 of God and His messenger, and 
they have succour, for whoever of them is just and pious (man 
barra wa-ttaqd}.+ 

2. The Messenger of God (God bless and preserve him) wrote 
to B. Ghifar that they are of the Muslims, with the privileges of 
the Muslims and the obligations of the Muslims; and that the 
prophet covenants to them the dhimmah of God and the dhimmah 
of His messenger, for their goods and persons ; succour is due to 
them against whoever begins wrong against them; when the 
prophet summons them to succour him, they are to respond to 
him; (incumbent) on them is his succour, 5 except (on) those who 
are fighting about religion, so long as the sea wets a piece of wool; 

1 i/2. 15-86; in the references the page and (sometimes) line of the standard 
edition is given, followed in brackets by the number of the paragraph in Well- 
hausen 's text and translation (Skizzen und Vorarbeiten, iv/3). 

2 This translation of a difficult passage seems to make sense without violence 
to the text; on this view the clause excludes from the duty of answering a 
summons those who are already engaged in an expedition 'on behalf of the 
religion of God*. Wellhausen differs. Cf. the similar passage in the following 
letter, and also Excursus G, I (j) and II (j). Another possible rendering is: 'the 
succour of the prophet is against them (sc. Pamrah) unless they fight for God's 
religion.' 3 i.e. guarantee of security, protection ; cf. p. 244 above. 

* IS, i/2. 27. 3-7 ( 40). 

5 i.e. giving succour to him; an alternative rendering is: 'against them is his 
succour, except him who fights for religion.' 


this writing does not come in front of (and protect from the 
penalties of) crime. 1 

3. The Messenger of God (God bless and preserve him) wrote 
to B. Zur'ah and B. ar-Rab'ah of Juhaynah that they have a 
guarantee for their persons and their goods; and that there is due 
to them succour against whoever wrongs them or fights them, 
except about religion and family (ahl) ; 2 there is due to the people 
of their nomadic part, who are just and pious, what is due to their 
settled part; and God is the one appealed to for help. 3 

4. The Messenger of God (God bless and preserve him) wrote 
to 'Amr b. Ma'bad al-Juham and B. al-Huraqah of Juhaynah and 
B. al-Jurmuz if any of them becomes a Muslim and performs 
the Worship and gives the zakat and obeys God and His messenger 
and gives of his spoils the fifth and the share of the prophet and the 
soft ('first pick'), 4 and if he professes Islam openly and keeps apart 
from the idolaters, then he is secure by the guarantee (amin bi- 
amdri) of God and the guarantee of Muhammad; where a debt is 
owing to any Muslim, (the repayment of) the capital is prescribed 
for him, but the interest on the pledge is void. The tax (sadaqah) 
on fruits is the tenth. He who joins them has the same rights as 
they have. 5 

5. The Messenger of God (God bless and preserve him) wrote 
to Budayl and Busr and the heads (sarawdt) of B. 'Amr. '. . . I have 
not betrayed your rights, nor injured your reputation. 6 The most 
honoured to me of the people of Tihamah, and the nearest of them 
to me in kin are yourselves, and those of the Mutayyabun 7 who 
follow you.' '. . . I have taken (or adopted, sc. as rights or privileges) 
for him of you who migrates (hdjara) the like of what I have taken 
for myself, even if he migrates in his own land, except the dweller 
in Mecca, apart from him who makes the lesser or greater pilgrim- 
age (m\itamir y hdjj). I have not done away with (anything) in respect 

1 26. 26 ff. ( 39). 

* Perhaps ahl ought to be omitted; the following word, translated 'people', 
is also ahl. 

3 24. 2-4 ( 27); Wellhausen thinks this and the two preceding treaties are 
with non-Muslims (p. 112 n.). 

4 Cf. p. 255 above. Add wa before af-$afi f 

5 24. 27-25. 5 ( 30 d). 

6 Or 'done away with what is due to you*. Wellhausen's translation would 
require *ald\ cf. instances of the meaning * instigate' in R. Dozy, Supplement aux 
Dictionnaires Arabes*, Leiden, 1927, ii. 

* Cf. MfMecca, 5 f., &c. 


of you since I made agreement. You have nothing to fear from 
my part and you are not oppressed.' 1 

6. The Messenger of God (God bless and preserve him) wrote 
to a mixed band (jumma), 2 who were in the mountainous part 
of Tihamah and had used violence on the passing (travellers), 
(the band being men) from Kinanah, Muzaynah, al-Hakam, and 
al-Qarah, together with slaves who followed them. When the Mes- 
senger of God (God bless and preserve him) triumphed, a deputa- 
tion came from them to the prophet (God bless and preserve him), 
and the Messenger of God (God bless and preserve him) wrote 
to them: 'In the name of God, the Merciful, the Compassionate. 
This is a writing from Muhammad the Prophet, the Messenger of 
God, to the servants of God, the freedmen ('utaqd') ; if they believe 
and perform the worship and give the zakat, then the slave among 
them is free (hurr) y and their patron is Muhammad ; he among them 
who is from a tribe is not handed back to it; whatever there is 
among them of blood they have shed or wealth they have taken is 
(given) to them (and blood- wit or repayment will not be exacted) ; 
whatever debt is (owing) to them among the people, will be repaid 
to them; there will be no wrong (done) to them and no hostility 
(shown); on these terms they have the dhimmah of God and the 
dhimmah of Muhammad. Peace be upon you. Ubayy b. Ka'b wrote 

7. There came before the Messenger of God (God bless and 
preserve him) the deputation of B. 'Abd b. 'Adi (of Kinanah), 
among them al-Harith b. Uhban, 'Uwaymir b. al-Akhram, and 
Habib and Rabf ah the sons of Mullah, and along with them 
a number of their tribe. They said, *O Muhammad, we are the 
people of the sacred area (haram ? of Mecca), and the dwellers 
there and the strongest of those in it; and we do not want to fight 
you; if you are fighting against others than Quraysh, we will fight 
with you, but we will not fight against Quraysh; we love you and 
those from whom you are ( ? Muslims) ; if you strike one of us by 
mistake, blood-wit for him is obligatory for you, and if we strike 
one of your companions, blood-wit for him is obligatory for us/ 
He said 'Yes', and they became Muslims. 4 

8. The first of (the tribal group of) Mudar to come as a deputa- 
tion to the Messenger of God (God bless and preserve him) were 

1 25. 11-16 ( 32 a). * Cf. p. 61 above, affair of Abu Ba?Ir. 

3 29. 13-22 ( 46). 4 48. 19-24 ( 91 b). 


400 men of Muzaynah. That was in Rajab (vii) of the year 5 
(December 626). The Messenger of God (God bless and preserve 
him) gave them (the dignity of) the hijrah in their own home. He 
said, 'You are Emigrants (muhajirUri), wherever you are; so return 
to your property (sc. herds)*. They returned to their country. 1 

9. The Messenger of God (God bless and preserve him) wrote : 
In the name of God, the Merciful, the Compassionate. From 
Muhammad the prophet to B. Asad. Peace be upon you. I praise 
God to you, besides whom there is no god. Furthermore, do not 
approach the watering-places and the land of Tayyi', for their 
watering-places are not lawful for you. Let no one enter their land 
except him whom they cause to enter. The dhimmah of Muham- 
mad is not responsible for him who disobeys him. Let Quda'i b. 
' Amr see to this. Khalid b. Sa'id wrote it. 2 

10. The Messenger of God (God bless and preserve him) wrote : 
In the name of God, the Merciful, the Compassionate ; these are (the 
terms) on which Nu'aym b. Mas'ud b. Rukhaylah al-Ashja f i made 
confederacy (hdlafa) ; he made confederacy with him on the basis 
of help and counsel (nasr, nasthah), so long as Uhud is in its place 
and so long as the sea wets a piece of wool. 'All wrote it. 3 

1 1 . (After the conversion of Rifa'ah b. Zayd of Judham, Muham- 
mad gave him a letter to his tribe.) 'This is a letter from Muham- 
mad, the Messenger of God, for Rifa'ah b. Zayd (to take) to his 
tribe, and whoever enters along with them ; he calls them to God ; 
whoever accepts is in the party of God (hizb Alldh) y and whoever 
rejects has safety for two months/ His tribe responded to him 
(favourably) and became Muslims. 4 

12. The Messenger of God (God bless and preserve him) wrote 
to whoever of Hadas of Lakhm became Muslim, performed the 
worship, paid the zakdt, gave the share (hazz) of God and the share 
of His messenger and separated from the idolaters he is secure 
(dmiri) by the dhimmah of God and the dhimmah of Muhammad; 
in the case of him who goes back from his religion, the dhimmah of 
God and the dhimmah of Muhammad, His messenger, is free of 
(responsibility towards) him; he whose isldm is attested by a 
Muslim is secure by the dhimmah of Muhammad, and is of the 
Muslims. 'Abdallah b. Zayd wrote it. 5 

1 38. 11-14 ( 76 a). a 23. 18-22 ( 24). 

3 26. 18-20 ( 35). 4 83. 2-4 ( 140 a). 

5 21. 14-19 ( 16). 


13. The Messenger of God (God bless and preserve him) wrote 
to Dughatir the bishop. 'Peace to him who believes. Furthermore, 
Jesus son of Mary is the spirit of God and His word; He placed 
it in Mary the pure. I believe in God and what was revealed 
to us and what was revealed to Abraham, Ishmael, Isaac and Jacob, 
and to the Israelites (asbdf), and what was given to Moses and 
Jesus, and what was given to the prophets from their Lord. We 
do not distinguish between any of them (sc. count some superior 
to others). We are surrendered (muslimuri) to Him. Peace be upon 
him who follows the guidance.' He sent this by Dihyah b. Khalifah 
al-Kalbl. 1 

14. The Messenger of God (God bless and preserve him) wrote 
to B. Janbah, who were Jews in Maqna, and to the people of 
Maqna. Maqna is a town of Aylah. 'Furthermore, your envoys 
have come to see me as they returned to your town. When this 
letter of mine comes to you, you are in security. You have the 
dhimmah of God and the dhimmah of His messenger. The Messen- 
ger of God pardons you your evil deeds, and all your faults. You 
have the dhimmah of God and the dhimmah of His messenger. To- 
wards you there is no wrong and no enmity. The Messenger of 
God is your neighbour (jar), (protecting you) from what he protects 
himself from. To the Messenger of God belong all the fine cloth 
and all the slaves among you, and all the horses and the armour, 
except what is not required by the Messenger of God or the 
messenger of the Messenger of God. In addition, there is due from 
you a quarter of what your palm-trees produce, a quarter of what 
your fishing-rafts (or fishermen) catch, and a quarter of what your 
women weave. Thereafter you are free from all tax (jizyah) or 
forced labour (sukhrah). If you hear and obey, it will be incumbent 
on the Messenger of God to honour the honourable man of you, 
and to pardon the wrong-doer. 

Furthermore, to the believers and Muslims (it is said) : He who 
does good to the people of Maqna, it will be good for him, and he 
who does evil to them, it will be evil for him. There is over you 
no ruler (amir) except from yourselves or from the people of the 
Messenger of God. Peace.' 2 

15. In the name of God, the Merciful, the Compassionate. From 
Muhammad the prophet to B. Zuhayr b. Uqaysh, a clan of 'Ukl. 

1 28. 6-1 1 (43). 

2 28. 12-23 ( 44) th e closing paragraph might be from a separate letter. 


If they confess that there is no god but God, and that Muhammad 
is His messenger, and keep apart from the idolaters, and agree to 
give the fifth of their spoils and the share of the prophet and his 
safl ('first pick'), then they are secure by the guarantee (amdri) of 
God and of His messenger. 1 

1 6. The Messenger of God (God bless and preserve him) wrote 
to the bishop of B. al-Harith b. Ka'b and the bishops of Najran 
and their priests and those who followed them and their monks, 
that for all their churches, services and monastic practices, few or 
many, they had the protection (jiwdr) of God and His messenger. 
No bishop will be moved from his episcopate, no monk from his 
monastic state, no priest from his priesthood. There will be no 
alteration of any right or authority or circumstance, so long as they 
are loyal and perform their obligations well, they not being bur- 
dened by wrong ( ? suffered) and not doing wrong. Written by 
al-Mughirah. 2 

17. This is a letter from Muhammad the prophet, the Messenger 
of God, to the people of Najran. To him belonged the decision 
upon them in respect of every fruit, yellow, white, or black, and 
every slave; but he was gracious to them and left (them) all that 
for the payment of 2,000 suits of clothes, namely, suits of ounces, 
of which a thousand are to be handed over each (year in) Rajab 
(vii) and a thousand in Safar (ii), each suit worth one ounce (Uqiyah). 
Where these tribute suits exceed or fall short of the ounce, that is 
taken into account. Whatever was taken from them of the coats of 
mail and horses and riding-camels and equipment they possessed 
is taken into account. Najran is to give lodging to my messengers 
for 20 days or less, but my messengers are not to be kept more 
than a month. It is obligatory for them to lend 30 coats of mail, 
30 horses, and 30 camels if there is war (kayd) in the Yemen. What- 
ever is destroyed of the coats of mail or horses or camels they lend 
my messengers is guaranteed by my messengers until they repay it. 
Najran and their followers have protection (jiwdr) of God and the 
dhimmah of Muhammad the prophet, the Messenger of God, for 
themselves, their community, their land, and their goods, both 
those who are absent and those who are present, and for their 
churches and services (no bishop will be moved from his episcopate, 
and no monk from his monastic position, and no church-warden 3 

1 30. 5-8 ( 4 8). 2 21 ( 14). 

3 Reading wdfih for wdqif with C. A. Nallino, Raccolta di Scritti, iii. 128 n. 


from his church-wardenship) and for all, great or little, that is 
under their hands. There is no usury, and no blood-revenge from 
pre-Islamic times (Id damm al-Jdhilfyah). If any of them asks for 
a right, justice is among them (sc. in their own hands) (to see that 
they are) not doing wrong and not suffering wrong; it belongs to 
Najran. If any one takes usury after this, my dhimmah is free from 
(responsibility for) him. No one of them is punished for the wrong- 
doing of another. On the terms stated in this document (they have) 
protection (jiwdr) of God and dhimmah of the prophet for ever, 
until God comes with His command, if they are loyal and perform 
their obligations well, not being burdened by wrong. Witnessed 
by Abu Sufyan b. Harb, Ghaylan b. ' Amr, Malik b. 'Awf an-Nasri, 
al-Aqra e b. Habis, al-Mustawrid b. 'Amr, a brother of Bali, al- 
Mughirah b. Shu'bah and 'Amir, client of Abu Bakr. 1 

1 8. The Messenger of God (God bless and preserve him) wrote 
to Khalid b. Dimad al-Azdi, that to him belongs the land he had 
when he became a Muslim, on condition that he believes in God 
He has no partner and confesses that Muhammad is His servant 
and His messenger, and on condition that he performs the Worship, 
pays the zakdt, fasts the month of Ramadan, makes pilgrimage to 
the house (sc. the Ka'bah), does not give shelter to any rebel (or 
disturber of the peace) and does not waver,' and on condition that 
he deals uprightly with God and His messenger and that he loves 
the friends of God and hates the enemies of God. On Muhammad 
the prophet it is incumbent to protect him (Khalid) and his pro- 
perty and family from what he protects himself from. Khalid al- 
Azdi has the dhimmah of God and the dhimmah of Muhammad the 
prophet, if he fulfils this. Ubayy wrote it. 2 

19. The Messenger of God (God bless and preserve him) wrote 
to al-Hilal, the master of al-Bahrayn : Peace to you. I praise to you 
God, besides whom there is no god, and who has no partner; and 
I summon you to God alone, that you believe in God, and obey, 
and enter into the community (jama ah). That is better for you. 
Peace be upon whoever follows the guidance. 3 

20. The Messenger of God (God bless and preserve him), on 
leaving al-Ji'ranah, sent al-'Ala' b. al-Hadrami to al-Mundhir b. 
Sawa al-'Abdi, who was in al-Bahrayn, summoning him to Islam, 
and he wrote a letter to him. He (al-Mundhir) wrote to the 

1 35 * ( 73); see also Abu Yuauf, Kharaj, 44 (tr. 108), and al-Baladhun, 63 f. 

2 IS, i/2. 21. 19-25 ( 17). 3 27, 7-10 ( 41). 


Messenger of God (God bless and preserve him) announcing his 
acceptance of Islam and his belief in him (or it) : 'I have read your 
letter to the people of Hajar. Some of them like Islam and admire 
it and have entered it; and some dislike it. In my country are 
Magians and Jews. Tell me your command about that.' 

The Messenger of God (God bless and preserve him) wrote to 
him : In so far as you act well, we shall not remove you from your 
position as ruler. He who remains a Jew or a Magian is obliged to 
pay the tax (jizyah). 

The Messenger of God (God bless and preserve him) wrote to the 
Magians of Hajar, presenting Islam to them. If they refuse, there is 
taken from them the jizyah' and their women are not to be married 
(by men of other religions) and their sacrifices are not to be eaten. 1 

21. The Messenger of God (God bless and preserve him) wrote 
to al-Mundhir b. Sawa. furthermore, my messengers have praised 
you. In so far as you act well, I shall act well towards you, and 
reward you for your work; and you shall deal uprightly with God 
and His messenger. Peace be upon you.' He sent it by al-'Ala b. 
al-Hadrami. 2 

1 19. 1-8 (9). 2 27. 25-28 ( 42). 


The Treaties with Dumat al-Jandal 

SEVERAL accounts have been preserved of negotiations between 
Muhammad and various groups among the inhabitants of Dumat 
al-Jandal. 1 Some of these require no special discussion, but two 
documents are worthy of detailed examination, since their authen- 
ticity has been questioned, while, if they are authentic or mainly 
authentic, they raise interesting questions. The text of the docu- 
ments may be given in full (divided into sentences to bring out the 
parallelism and for convenience of reference). 2 

(a) This is a letter from Muhammad, the Messenger of God, to 
Ukaydir, when he agreed to become a Muslim (or 'to submit', sc. to 
God and to Muhammad) and repudiated the gods and idols, along with 
Khalid b. al-Walid, the Sword of God, in respect of Dumat al-Jandal 
and its neighbourhood (or * removed the gods . . . which were 
in . . .'). 

(b) To him (sc. Muhammad) belongs the outer part where there is 
little water, what is uncultivated, what is not marked off and what is 
not appropriated of the land ; and the armour, the arms, the horses and 
the stronghold. 

(c) To you belong the inner palm-trees, and the cultivated land 
watered by springs ( = or 'appropriated'). 

(d) And far distant is the fifth (meaning that the fifth is not to be 
exacted; or 'and after the fifth', meaning that it is to be paid before 
possession is established, or before the following clause applies). 

(/) Your pasturing beast is not to be turned away (from the pasture ; 
? before tithing) ; your segregated beast is not to be reckoned (as liable 
for tithe). 

(g) You are not to be debarred from the pasture (lit. 'plants', pre- 
sumably in the outer area). 

(h) The tithe is not to be taken from you except on the established 
(thabdt, sc. palms). 3 

(i) You are to perform the Worship (saldt) at its (proper) time, and 
to give the zakdt as is due. 

1 IS, i/2. 34 ( 66), 36 ( 73), 68 f. ( 119); cf. Caetani, ii/i. 259-70- 

2 IS, i/2. 36, 69; cf. IH, 903 (and ii. 205); WW, 403 f.; Tab. 1702 f.; al- 
Baladhuri, 61 ff. 3 IH, ii. 205 has v. 1. nabdt, plants, as in (g). 


(j ) On those terms you have an agreement and covenant, and thereby 
you have (to expect from us) uprightness and performance of duties. 
(k) Witnessed by God and those of the Muslims who were present. 


(a) This is a letter from Muhammad, the Messenger of God, to the 
people of Dumat al-Jandal and those who follow them of the clans 
(tawd'if) of Kalb, along with ( ? sent by) Harithah b. Qatan. 

(b) To us belongs the outer rain-watered (or unwatered) part. 

(c) To you belong the inner palm-trees. 

(e) Where there is running water a tenth is due; where there is 
spring- water half a tenth is due. 

(/) Your pasturing beasts are not to be joined together; your 
segregated beasts are not to be turned away. 

(i) You are to perform the Worship at its (proper) time, and to give 
the zakdt as is due. 

(g) You are not to be debarred from the pasture. 

(h) The tithe is not to be taken from you on household goods (batdt). 

(j) On those terms you have an agreement and covenant, and we owe 
you ( ? you owe us) sincerity and fulfilment, and the dhimmah of God 
and His messenger. 

(k) Witnessed by God and those of the Muslims who were present. 

The first comment to be made on these documents is that there 
are several places where the text is uncertain and the meaning 
obscure. The most notable is (h) where there are three readings, 
thabdt, nabdt, and batdt, which could easily be confused with one 
another in Arabic script; moreover the word* except* (ilia) seems to 
have fallen out in II. The precise meaning of clauses (/) and (g) is 
also obscure, but it is not specially relevant to the present discussion. 

The next point to notice is that there are no cogent reasons for 
thinking that I is a later invention based on II. Ukaydir is a well- 
attested historical character. Our ignorance of his relation to the 
other inhabitants of Dumat al-Jandal and its neighbourhood is not 
a ground for denying his existence. There was presumably room 
for several distinct groups. It may well be, as Caetani suggests at 
one point, that Ukaydir (who was from the South Arabian tribe of 
Kindah) ruled over the town-dwelling immigrants from 'Iraq, 
while the tribesmen of Kalb were partly agricultural and partly 
nomadic; the distinction cannot have been sharp, however, since 
Ukaydir's letter speaks of herds and Harithah's of palms. Once the 
existence of different groups in the oasis is admitted, there is no 


improbability in two letters being similar. Moreover, such a varia- 
tion as the addition of the closing words of (b) in I is appropriate 
to what we know of Ukaydir's situation. 

The question whether Ukaydir and Harithah's group are Chris- 
tians or Muslims leads us to the heart of the problem. There is 
evidence elsewhere that Ukaydir remained a Christian. 1 The tenor 
of the documents, too, implies the surrender of non-Muslims to 
Muslims. Note the contrast between 'us* and 'you' in II (b, c) and 
the mention of 'the Muslims' in (k). The case of Hawazin, who 
became Muslims and yet had to make a liberation-payment 
(si* ayah), 2 is not an exact parallel ; had there been something similar 
at Dumat al-Jandal we should presumably have heard more about 
it. It is almost certain, therefore, that the persons to whom these 
letters were written were not Muslims ; and we have next to explain 
how the texts contain passages which seem to imply that they were 
(i) and I (a). 

There are two possible explanations. Firstly it might be held 
that (a) in I is a later editorial heading, and that (/) has also been 
added in both in different positions, be it noted at some later 
period; the phrase 'repudiated the gods and idols' is not altogether 
suitable for a Christian. Secondly it might be held that isldm in I (a) 
does not have the technical sense of 'becoming a Muslim', but 
means 'submitting' merely; again, 'performing the Worship' might 
perhaps refer to Christian rites and zakdt might be applied to the 
payments made to Muhammad by monotheistic subject groups. 

Of these two possibilities the former is the more probable, but 
it is not necessary here to make a final decision between them. On 
either explanation the problem remains. These persons (we have 
argued) are not Muslims ; yet they have to make payments which 
are identical in principle and in manner of calculation with the zakdt 
paid by Muslims. Indeed, document II appears in Ibn al-Athir's 
Usd al-Ghdbah 3 reduced to clause (e) ; and this suggests that tliis 
document was an important source for the principle that the zakdt 
on naturally and artificially irrigated land is a tithe and a half-tithe 
respectively. There is here a potent reason for the insertion of 
(i) and I (a). These people were paying something that was materi- 
ally identical with zakdt; therefore, the early scholars would argue, 
they must have been Muslims. Finally, if this train of reasoning is 

1 e.g. mention ofjizyah in IH, 903 and WW, 404. 2 Cf. p. 101 above. 
3 * 357, s.v. #rithah b. Qajan; cf. al-Bukhri, Zakdt (24), 55. 


sound, it leads to an important conclusion, namely, that for a time 
presumably before the revelation of the verse about jizyah* or 
at least before its general application some allied or subject non- 
Muslim groups made payments to Medina that were identical with 
those made by Muslim groups. Such a conclusion, of course, is at 
the same time the setting of a problem for future research. 

1 9. 29; cf. pp. 115, 255 above. 


List of Administrators sent out by Muhammad 

(This list merely contains the main references from Ibn Hisham 
and at-Tabari, along with one or two others, and does not claim 
to be complete.) 

(a) Sent to the south 

1. al-Muhajir b. Abi Umayyah (Quraysh-Makhzum) ; sent to 
San'a' (IH, 965 = Tab. i. 1750; cf. IH, 971 foot); to B. Mu e - 
awiyah b. Kindah, but did not go until the caliphate of Abu Bakr 
(Tab. 1853). 

2. Ziyad b. Labid (Ansar-Bayadah) ; to Hadramawt (IH, 965 = 
Tab. 1750); was also set over the district of al-Muhajir (B. 
Mu'awiyah?) (Tab. 1852 f.). 

3. 'All b. Abi Talib (Quraysh-Hashim) ; sent to Najran to collect 
sadaqat andjizyah (IH, 965 = Tab. 1750). 

4. Mu'adh b. Jabal (Ansar-Salimah) ; to Himyar (IH, 956); to 
the Yemen (IS, i/2. 20) ; to teach in the Yemen and Hadramawt 
in districts that were under various 'agents', 'ummal (Tab. 
1852 f.). 

5. 'Abdallah b. Zayd (Ansar-Ba '1-Harith); to Himyar (IH, 956; 
? also in Tab. 1853). 

6. Malik b. 'Ubadah (Hamdan); sent to Himyar (IH, 956). 

7. 'Uqbah b. Namir (Hamdan); to Himyar (IH, 956). 

8. Malik b. Murrah (or Murarah) ar-Ruhawi (Madhhij); to 
Ilimyar (IH, 956; cf. IS, i/2. 20). 

9. Shahr b. Badham (Persian) ; recognized as governor of San c a* 
(Tab. 1852 f.). 

10. e Amir b. Shahr al-Hamdam; to Hamdan (Tab. 1852). 

11. Abu Musa ('Abdallah b. Qays) al-Ash f ari; to Ma'rib (Tab. 

12. Khalid b. Sa'id b. al-'As (Quraysh-'Abd Shams); to region 
between Najran, Rima c and Zabid (Tab. 1852; cf. al-Ya'qubi, 
ii. 81, to San'a'). 

13. at-Tahir b. Abi Halah (Tamim, confederate of 'Abd ad-Dar 
of Quraysh); to 'Akk and Ash'ar (Tab. 1852). 

14. Ya'la b. Umayyah (Tamim, confederate of Nawfal of 
Quraysh); to al-Janad (Tab. 1852). 


15. 'Amr b. IJazm (Ansar-an-Najjar); to Najran (Tab. 1852); 
sent to the Yemen (perhaps to B. al-Harith b. Ka'b) (IH, 961 f. = 
IS, i/2. 2). 

16. 'Ukkashah b. Thawr al-Ghawthi; to Sakasik, Sakun and (?) 
Mu'awiyah b. Kindah (Tab. 1852 f.). (N.B. Most of the names 
mentioned in Tab, 1852 f. also occur on 1952 f.). 

17. Abu Sufyan b. Harb (Quraysh-'Abd Shams); in Jurash 
of the Yemen (al-Baladhuri, 59); in Najran (Usd, s.v.; p. 75 

(b) To the east 

18. al-'Ala' b. al-Hadrami (confederate of Quraysh 'Abd 
Shams); to al-Bahrayn (IH, 945, 965; Tab. 1750; IS, i/2. 19; 
cf. al-Ya'qubl, ii. 81, al-Ghutayf bi '1-Bahrayn). 

19. al-Aqra' (?b. Habis) (Tamim); musaddiq for Hajar (al- 

(c) In the neighbourhood of Medina 

20. al-Walid b. 'Uqbah b. Abi Mu'ayt (Quraysh-'Abd Shams); 
to al-Mustaliq, repulsed (IH, 730 f. ; WW, 387). 

21. e Adi b. Hatim (Tayyi') ; to Tayyi* and Asad (IH, 965 = Tab. 


22. Malik b. Nuwayrah (Tamim) ; to collect sadaqat of B. Han- 
zalah of Tamim (IH, 965 = Tab. 1750). 

23. az-Zibriqan b. Badr (Tamim); to half of B. Sa f d of Tamim 
(IH, 965); cf. p. 139 above. 

24. Qays b. 'Asim (Tamim-Minqar) ; to half of B. Sa c d of Tamim 
(IH, 965); cf. p. 139 above. 

25. Buraydah b. al-Husayb (Aslam) ; collected tax of Aslam and 
Ghifar (WW, 385); summoned Aslam for Tabuk (ibid. 391). 

26. Ka'b b. Malik (Ansar-Salimah) ; variant for above (ibid. 


27. 'Abbad b. Bishr (Ansar-'Abd al-Ash'hal); collected tax of 
Sulaym and Muzaynah (WW, 385); and of al-Mustaliq after al- 
Walid b. 'Uqbah (ibid. 387); summoned Aslam (ibid. 391). 

28. Rafi f b. Makith (Juhaynah) collected tax of Juhaynah (WW, 
385); summoned tribe along with brother (ibid. 326, 391). 

29. c Amr b. al-'As (Quraysh-Sahm) ; tax of Fazarah (WW, 

30. ad-Dahhak b. Sufyan (Kilab); tax of Kilab (WW, 385). 


31. Busr b. Sufyan (Ka'b); tax of Ka'b (WW, 385); summoned 
Ka'b, along with Budayl b. Warqa' and 'Amr b. Salim (ibid. 326, 

32. Nu e aym b. 'Abdallah an-Nahham (Quraysh-'Adi) ; variant 
for collector from Ka'b (WW, 385). 

33. Ibn al-Lutblyah (Azd); tax of Dhubyan (WW, 385). 

34. ' Uyaynah b. Hisn (Fazarah) : sent to collect tax of Tamfm 
in April-May 630 = i/g (IS, ii/i. 115. 19, presumably referring to 
the expedition to al- f Arj ; the list in WW, 385 is repeated here). 

35. Quda'ib. c Amr ( c Udhrah); over Asad, &c. (IS, i/2. 23. 22 f. ; 
cf. p. 357 above). 

Zakdt and Sadaqah 

As is well known sadaqah in later Islamic usage commonly means 
'voluntary alms', while zakdt means the prescribed 'legal alms', 
whose amount is fixed, though in practice the giving of it may be 
voluntary. It is in this sense that zakdt is one of the five 'pillars of 
Islam'. The use of the word zakat in the Qur'an, however, raises 
serious difficulties. There are thirty-one instances of it, and in 
nearly all of them we have some form of the phrase * observing the 
Worship and giving the zakat'. 1 This is evidently, then, a technical 
phrase. We should consequently expect to find it used chiefly 
towards the end of the Medinan period, when the nomadic tribes 
were making treaties with Muhammad, and he was fixing the 
amount of zakat for them. According to Richard Bell's dating, 
however, though all the uses of the word are Medinan (with the 
possible exception of 18. 81/80 and 19. 13/14 where the meaning 
is different in any case), several of them fall early in the Medinan 
period. How is this to be explained? Was the zakat exactly 
prescribed at an early time ? Or was the word zakat used in the 
meaning of 'voluntary contribution' ? If the latter, was there any 
distinction from nafaqah, ma tunfiqu, and the like, which is 
commonly translated 'contribution' ? 

Part of the explanation is perhaps to be found in the fact that 
many of the earliest passages where zakat occurs refer to the Jews 
or to disaffected Medinan Arabs, who may be presumed to have 
been friendly with the Jews. The point is sufficiently important 
for it to be illustrated in detail. 

7. 156/155 E . In reply to Moses at Sinai, God promises mercy 
to those who are pious, give the zakat and believe in His signs. 

5. 12/15 F. God (at Sinai) made a covenant with the Israelites 
and promised a reward if they observed the Worship, gave the 
zakdt, believed and supported His messengers and lent to 
Him a good loan. 

2. 83/77 FG - God made a covenant with the Israelites and 

1 Aqdmu 9 f-faldt wa-dtaw 'z-zakdt. Exceptions: 18. 81/80; 19. 13/14, where 
the meaning is probably 'purity*; 23. 4 has perhaps the same meaning, but $aldt 
is mentioned in the context; 7. 156/155; 30. 39/38; 41. 7/6. 

5788 B b 


commanded them, among other things, to observe the 
Worship and give the zakdt. 

21. 73 EL God revealed to Isaac and Jacob the doing of good, 
observing of the Worship and giving of zakdt. (Cf. 19. 31/32, 
55/56 EI where Jesus is charged with the Worship and zakdt 
(?for himself), and Ishmael enjoins them on his people.) 

98. 5/4 F. The people of the Book were commanded to serve 
God alone, as hanifs, to observe the Worship and give the 

2. 43/40 ? F. An appeal to the Jews to observe the Worship and 
give zakdt. 

4. 162/160 ? GH. Some Jews believe in God and the Last Day, 
observe the Worship and give zakdt] they will be rewarded. 

30. 39/38 EI. Zakdt is rewarded by God, but not money given 
for usury (the reference to usury is probably to the Jews). 1 

24. 37 ? H. Men who are not diverted by trade and bargaining 
from remembering God, observing the Worship and giving 
zakdt. (Bell apparently interprets of the Meccan merchants 
and thinks that the mention of Worship and zakdt was added 
later; but it might refer to the Jews and be early Medinan, 
especially as places of worship are mentioned in the previous 

2. 1 10/104 ? FG. An exhortation to observe the Worship and give 
zakdt. (This occurs as part of a warning against Jewish infl i- 
ence, and hence is primarily addressed to friends of the Jews.) 

24. 56/55 G. A command to observe the Worship, give the zakdt 
and obey Muhammad. (This occurs in a passage addressed to 
disaffected persons, but does not fit the context, and was 
perhaps revealed separately.) 

22. 41/42 ? G. God will support those who, among other things, 
observe the Worship and give zakdt. 

5. 55/60 FG. The believers are to be friends (not with the Jews) 
but with God, His messenger, and those who believe, observe 
the Worship and give zakdt (presumably addressed to friends 
of the Jews). 

4. 77/79 ? E (or G). Some of those who were told to restrain 
their hands, observe the Worship and give the zakdt y are 
afraid of fighting. (Presumably friends of the Jews ; the sur- 
rounding verses are from the time after Uhud, and suggest 
1 Cf. p. 297 above. 


that this verse is of the same date, but Bell thinks it is possibly 
early Medinan.) 

Of the remaining usages of the word six (9. 5, n, 18, 71/72; 
33- 3J 58. 13/14) are definitely late. One (23. 4E) is dated early 
Medinan, but might conceivably have the meaning of 'purity', 
though 'the Worship* occurs in a neighbouring verse (according 
to a tradition this was the last surah revealed before Muhammad 
left Mecca). Two (2. 177/172 F ; 22. 78 F) seem to be about the 
time of the change of qiblah y which is perhaps significant; in the 
first Bell thinks the mention of zakdt has been added by way of 
revision, but this may be an unnecessary supposition in view of the 
connexion with the Jews we are now studying. Four others (2. 
276/277; 31. 4/3; 41. 7/6; 73. 20) are described generally as 
'Medinan', and may or may not be early. 

The results of this examination are most surprising. One would 
have expected zakdt to acquire its 'technical' sense in the late 
Medinan period, when alliances were being formed with nomadic 
tribes. But only six instances clearly belong to this period. A larger 
number belong to the early period when Muhammad was in close 
touch with the Jews and trying to gain their support. The concep- 
tion of zakdt appears to have been adopted and developed in the 
Qur'an because it was already familiar to the Jews 1 and to those 
who had been influenced by them. That seems to be the point of 
the references to the covenant at Sinai. Muhammad must have 
been insisting that the Jews should give zakdt (whatever that may 
have implied). The Qur'an supports his demand by showing that 
it is in accordance with what God had previously demanded of 
them and they had agreed to. 

In the light of this strong connexion of zakdt with the Jews, it 
is probable that the cognate word yuzakki, used of a messenger 
towards a people, 2 means 'to appoint zakdt for', though the thought 
of it as a means of purification is not necessarily absent. 3 

An examination of the use of sadaqah (plural sadaqdt) in the 
Qur'an leads to a similar result. There are not so many instances 
as of zakdt, and the word is not coupled with 'the Worship'. One 
instance (2. 276/277 ? E) contrasts sadaqdt with usury, 4 in the same 

1 Cf. its probable origin in Aramaic; Jeffery, Vocabulary, &c. 

2 Cf. M\Mecca y 165, 'second group*. 

3 Cf. IS, iii/i. 93. 1 6 works of charity by 'Abd ar-Rahman b. 'Awf will be 
a 'purification of what he has 1 (tazkiyat md huwafi-hi). 

4 Cf. also 2. 280 E. 


way as zakat had been contrasted. The other four early instances 
(2. 263/265, 264/266, 271/273 all ? FG; 4. 1 14 GH), with the possible 
exception of the third, have the Jews or hypocrites in view. There 
are also some later instances. In Muhammad's letters and treaties 
reproduced by Ibn Sa c d (i/2) sadaqah is preferred to zakat for what 
may now be called 'legal alms'. 

There remains much that is obscure. Why should the Qur'an 
sometimes speak of az-zakat, sometimes of sadaqah, and sometimes 
of nafaqah or ma anfaqtum ? Was there any difference (in the early 
Medinan period) in the thing to which they refer, or are they 
merely directed to different groups of people ? It might be sug- 
gested that nafaqah, &c. were contributions to war purposes ; but 
the passage 2. 261/263 ff. identifies sadaqat and contributions, and 
speaks of sadaqat being given secretly; 2. 215/211 says that contri- 
butions are for parents, relatives, the poor, &c. Thus there is no 
evidence of any fixed levy on any section of Muhammad's followers. 
Those who had something to spare were required to help those 
in need, and presumably most of the Emigrants were in this cate- 
gory at first. If for the early period zakat is identified with sadaqah, 
there would have been no tax on the Jews, but they would have 
been expected to help people in difficulties by gifts, and not by 
lending money at interest. 1 

1 Cf. p. 297 above, esp. n. 4. 


Marriage and the Family in pre-Islamic Times 

NINETEENTH-CENTURY scholars, notably W. Robertson Smith 
and G. A. Wilken, following the anthropological ideas of their day, 
explained the phenomena of Muhammad's Arabia by the hypo- 
thesis of an age in which matriarchy was dominant. It is now widely 
recognized, however, that to posit some previous historical situa- 
tion for which there is no direct evidence does little to explain the 
observed or recorded facts of social anthropology, and that it is 
better to analyse the phenomena with a view to discovering the 
structural or functional principles actually present. For pre-Islamic 
Arabia there is a great mass of material available, though most of 
it is found only in Arabic. Most of it, too, can only be satisfactorily 
interpreted by someone well versed in anthropology. A thorough 
examination of the material is clearly out of place in a biography 
of Muhammad. In the body of this work I have said only enough 
to make possible an intelligible account of the reform of the system, 
while the present Excursus sets out some facts which justify what 
was said above. 

(a) Patrilineal features 

In most primitive societies both patrilineal and matrilineal prin- 
ciples are to be found at work. In pre-Islamic Arabia Mecca is the 
clearest example of a system that is predominantly patrilineal. We 
have to remember, of course, that our material was not written 
down till at least a century after Muhammad's death, and that in 
the interval the patrilineal system had ousted most traces of the 
matrilineal from Islamic society, so that the men who wrote down 
the material may have misunderstood some of the old practices they 
tried to describe. Nevertheless patrilineal relationships must have 
been important in Mecca. For one thing the chief social units were 
patrilineal clans, named after male ancestors only. Moreover, from 
about the time of Qusayy at least, these ancestors appear to be real 
figures and not mere names in a genealogy. This is in contrast to 
Medina where, until shortly before the Hijrah, the genealogies 
consist of names about which nothing is known. In accordance 
with this naming of the clans in Mecca, both men and women are 


known there almost exclusively by paternal descent, that is, as the 
sons and daughters of males, not of females. In the kunyah, too, 
at Mecca the honorific title, 'father of X* or 'mother of Y* 
(Abu X, Umm Y) X and Y are always sons. 

Among the exceptions is the name sometimes given to Muham- 
mad, Ibn Abi Kabshah, but it is the kind of exception which proves 
the rule. Kabshah was a common feminine name at Medina, as 
may be seen from Ibn Sa'd (especially iii/2, viii, and the Index), 
but apparently does not occur at Mecca. As there are other 
instances at Medina of a kunyah from a female (for example, Abu 
Lubabah), the probability is that this name is connected with 
Medina. There is nothing to show that it indicates any relationship 
with Muhammad's freedman, Abu Kabshah, who fought as a 
Muslim at Badr. The use of this name for Muhammad would 
mean, then, that either his father or one of his ancestors through 
his grandfather's mother had the kunyah Abu Kabshah. The rude- 
ness in the use of the appellation by Abu Uhayhah and other oppo- 
nents (if there was any) would consist in reminding Muhammad 
of his connexion with Medina and its queer ways. 1 A tribesman, 
presumably from a matrilineal background, seems to have used it 
with a perfect courtesy. 2 

There are also one or two examples among Quraysh of men 
being known by their mothers' names. 'All had a son known as 
Muhammad b. al-Hanafiyah, but this was probably to distinguish 
him from two other sons of 'AH also called Muhammad. 3 An 
example that cannot thus be explained away is that of Muhammad's 
opponent, Abu Jahl, who was sometimes known as Ibn al-Han- 
zaliyah after his mother, Asma' bint Mukharribah of the tribe of 
Hanzalah, a branch of Tamim. 4 Asma' later appears as a merchant 
in perfumes on her own account, and that further supports the 
belief that matrilineal ideas were still strong in this household. 
Moreover, Umm al-Julas bint Mukharribah, described as khdlah 
(maternal aunt) of Abu Jahl, is sometimes said to have been the 
person with whom the written agreement for the boycotting of 
B. Hashim was deposited; 5 but elsewhere Umm al-Julas is said to 
have been the kunyah of Asma'. (There is also confusion between 

1 IS, iv/i. 69. 3; WW, 48, 137; Tab. i. 1565; M/Mecca, 103. 

2 Ibid. i/2. 145. 27. 3 Ibid, iii/i. n f. 

4 IH, 441 ; IS, viii. 220; Usd, v. 393 f., s.v. Asma* bint Salamah (sic); WW, 61. 

5 IS, i/i. 40. i. 


the mother of Abu Jahl and her niece, married to her son 'Ayyash, 
Asma' bint Salamah b. Mukharribah; and in some texts Muk- 
harribah is corrupted to Makhramah.) 

Patrilineal features can also be observed in connexion with in- 
heritance, blood-revenge, and the place of marriage. Inheritance 
in the male line seems to have been common at Mecca, though 
without any rule of primogeniture. The power of Qusayy passed 
to his sons, grandsons, and so on; and it is to be presumed that to 
some extent this succession to power was based on inherited wealth. 
Similarly in the clan of Makhzum the power of al-Mughirah went 
to his son al-Walid and then to the latter's nephew Abu Jahl. Again, 
the duty of avenging blood lay primarily on the next-of-kin in the 
male line. This was perhaps the case even in groups that were 
mainly matrilineal. The son of Kulayb, the great chief of Taghlib, 
avenged his father's death by killing his mother's brother, Jassas; 
but the close proximity of Jassas and Kulayb shows that Kulayb 
had gone to live with his wife's people, whose property was under 
the control of Jassas, and so the system may have been mainly 
matrilineal. 1 The groups responsible for blood-money according 
to the Constitution of Medina appear to have had a patrilineal basis. 

About the place of residence of husband and wife there is little 
clear information for Mecca. The impression that marriage was 
often virilocal is supported by names like 'the shib (quarter) of 
Abu Talib', 'the house of al-Arqam 1 , and by the fact that some of 
the Muslims who emigrated to Abyssinia were accompanied by 
their wives (not all of whom can have been accompanying male 
relatives). There are also several cases of uxorilocal marriage, though 
of course this is a feature that may occur in mainly patrilineal 
societies. Muhammad's mother remained among her own kin, and 
'Abdallah merely visited her there; Muhammad was with her, 
apart from his time in the desert, until her death, and only then 
went to the house of his paternal grandfather, ' Abd al-Muttalib. 2 
It is interesting that in connexion with Muhammad's birth it was 
said that there was 'prophethood in the clan of Zuhrah' (his 
mother's clan); from the fact that the first transmitters of this 
tradition belonged to Zuhrah it may be inferred that, apart from 
the practice of uxorilocality, matrilineal ideas were strong in this 
clan. 3 Muhammad's own marriage with Khadijah was presumably 

1 Caussin de Perceval, ii. 336; Aghdni, iv. 150 f.; Robertson Smith, Kinship, 
155 f. * IS, i/i. 58. 17, 74- 3 Ibid. 51. 23. 


uxorilocal. Khadijah must have been a woman with some power 
and wealth, and may have married Muhammad on her own initia- 
tive. There are stories about her father or her uncle giving her in 
marriage, but the discrepancies in the stories make it probable 
that neither did so, and that the stories reflect later practice. 1 How 
Khadijah came to have her special position is very obscure. Apart 
from the growth of individualism it would presumably have been 
impossible for Khadijah and Asma' bint Mukharribah to trade in 
their own name. 

In general, as we shall see, there appears to be a connexion 
between the growth of individualism and the extension of the 
patrilineal system. Yet there are some pre-Islamic practices in 
which the emphasis on paternity is associated not with individual- 
ism but with the unity of the group of agnate males. The essential 
unity of a man and his consanguine brother (and presumably, by 
extension, of the whole patrician) is expressed in an Arab saying, 
'the paternal uncle is as the sinw of the father'. The word sinw 
means one of a pair or triad, and is used especially of palms 
growing from a single root. It can mean either consanguine or 
uterine brother, or even son. The application to palm-trees (which 
may be the original use) shows that the word emphasizes the unity 
of stock. 2 

Among the practices based on this idea is the marriage of 
brothers to the same woman, of which there dre several examples 
from Medina. 3 It is presumably a form of the levirate. There is also 
at least one instance from Mecca, Asma' bint Mukharribah, the 
mother of Abu Jahl, who married Abu JahFs fatHer Hisham b. 
al-Mughirah and also the latter's brother Abu Rabf ah. 4 There 

1 IS, i/i. 58 f.; cf. Robertson Smith, 274 f. and 99. 

2 Lane, s.v.\ cf. Ibn Hajar al-'Asqalam, Tahdhtb at-Tahdhlb y Hyderabad, 

(i9O7)/i325, &c., viii. 119 on 'Amr b. Yahya b. 'Amrnarah. 

3 Hind bint Simak (Bali, confederates of Qawaqilah) Sa'd and Aws bb. 
Mu'adh ('Abd al-Ash'hal); IS, iii/2. 99. 16, viii. 231. 

Suhaymah (Waqif) Salamah and Rumi bb. Waqsh ('Abd al-Ash'hal); IS, 

viii. 235. 19 f., 236. 7 f. 
Buraydah bint Bishr (Zafar) 'Abbad and Abu Ma'qil bb. Nahik (Harithah); 

IS, viii. 251. 
An-Nawar bint Qays (IJarithah) ayfi and Zayd bb. 'Amr (Kterithah); IS, 

viii. 240. 11-14. 
Umm al-Harith bint Malik (Salimah) Jabbar and Thabit bb. Sakhr (Sali- 

mah); IS, viii. 292. 23, 297. 5 f. 

Layla bint Abl Sufyan (Pubay'ah) Mu'adh and Bakir bb. 'Amir b. jSriyah? 
(pubay'ah); IS, 253. 6-10. 

(Cf. G. H. Stern, Marriage in Early Islam, 172 f.) 4 IS, viii. 220. 


are also examples of the marriage of father and son to the same 
woman, that is, of a man to his stepmother; and these are from 
Mecca, Medina and elsewhere. 1 The underlying idea was perhaps 
that the woman's child-bearing capacity belonged to this patrician 
(a suitable compensation having been given to her own patrician), 
and the son or brother of the deceased has the right or duty of 
actualizing the potentialities. 2 Muhammad b. Ka'b al-Qurazi (d. 
c. 736/118) is reported to have said that 'when a man died leaving 
a widow, his son was the person with the best right to marry her 
(ahaqq bi-hd an yankiha-ha, presumably not yunakkiha), if he 
wanted to do so, provided she was not his mother'. 3 

Another obscure matter may be mentioned at this point, since 
there seems to have been a difference between Mecca and Medina. 
This is the attitude towards ghllah or ghayl, that is, having sexual 
intercourse with a woman while she is suckling a child. The Mec- 
cans apparently thought that this was harmful to the child and 
objected to the practice, but there does not seem to have been the 
same objection to it at Medina. 4 There is a tradition to the effect 
that Muhammad had intended to forbid al-ghtlah until he remem- 
bered that the Persians and Greeks practised it without injury to 
their children. 5 Despite this tradition, however, Muhammad did 

1 Sukhta bint Harithah (Sa'idah) al-Mundhir b. Haram, Thabit b. 
al-Mundhir (Maghalah) ; IS, iii/2. 63. 

Hind bint Aws (Khatmah) an-Nu'man b. Umayyah, Thabit b. an-Nu'man 

('Amr b. *Awf); IS, iii/2. 44. 25, 45. 5. 
Mulaykah bint Kharijah (a wife of 'All) married a Fazari and his son ; Robertson 

Smith, 89, 271 ; cf. Aghdm, xxi. 261 (quoted by Farrukh, Das Bild des 

Fruhislam, 116); the names are variously spelt. 
'Aminah (mother of Abu Mu'ayt) married Umayyah b. 'Abd Shams and his 

son Abu 'Amr; Robertson Smith, 89. 
Nufayl (grandfather of 'Umar) left a Fahmi widow who was married by his 

son; Robertson Smith, 89. 
Kubayshah bint Ma'n Abu Qays, Hin b. Abi Qays; Robertson Smith, 271 

from al-Wahidl on Q. 4. 26. 
Umm 'Ubayd bint Ipamrah al-Aslat, Abu Qays b. al-Aslat ; Robertson Smith, 

271 from at-Tabari on Q. 4. 26. 
Bint Abi Talhah ('Abd ad-Dar) Khalaf, al-Aswad b. Khalaf (JumatO; 

Robertson Smith, 271 f. from al-Wahidl and at-Tabari. 
Fakhitah bint al-Aswad (Asad) Umayyah b. Khalaf, afwan b. Umayyah 

(Jumah); Robertson Smith, 271 f. from the same sources. 

2 Cf. A. R. Radcliffe-Brown and C. Daryll Forde, African Systems of Kinship 
and Marriage, London, 1950, 50 f., &c.; Robertson Smith, 87. 

3 IS, iv/2. 95. 27 f. 

4 G. H. Stern, Marriage in Early Islam, 96 f. : Lane, s.v. ghilah. 

5 Lane, ibid., IS, viii. 177. 16. 


not consummate his marriage with Umm Salamah until she had 
stopped suckling Zaynab, her daughter by Abu Salamah; 1 and 
another woman refused to marry Muhammad because she had two 
children by a former husband to nurse. 2 The examples are of 
avoiding intercourse with a woman while she is suckling another 
man's child. It is not clear, however, what the underlying idea is 
and whether there would have been the same objection when the 
woman was suckling the man's own child or his brother's child. 
After the Hijrah there was a strong feeling against intercourse with 
a woman pregnant by another man, and it was spoken of as 'irrigat- 
ing another man's crop'; but whether there was much of this 
feeling before the Hijrah we cannot tell. 3 

(b) Matrilineal features 

Despite the dominance of patrilineal ideas at the time when our 
sources were written down, many points of matrilineal organization 
have been recorded. Descent in the female line was relatively more 
important at Medina than at Mecca. Some clans, such as Banu 
Hudaylah and Banu Maghalah, took the name of a woman; and 
the Aws and the Khazraj together, before they became the Ansar 
on conversion to Islam, could be called Banu Qaylah after a 
common ancestress. Individuals also were known by their mothers. 
Among the best-known examples are Mu'adh, Mu'awwidh, and 
'Awf, the sons of 'Afra', while 'Abdallah b. Ubayy is sometimes 
called Ibn Salul after Ubayy's mother. An interesting case is that 
of Ka'b b. al-Ashraf, who was reckoned as belonging to his Jewish 
mother's clan of an-Nadir, although his father was of a nomadic 
Arab tribe. 4 Frequently, too, in dealing with the Medinan 'com- 
panions', Ibn Sa'd remarks that two men were ibn khdlah of one 
another, that is, sons of sisters; e.g. 'Abdallah b. Ubayy and Abu 
'Amir ar-Rahib, Sa e d b. Mu'adh and As'ad b. Zurarah. 

As a prelude to a discussion of the form of the marriage relation- 
ship and the place of residence of the spouses, it will be convenient 
to quote the account of pre-Islamic marriage recorded by al- 
Bukhari : s 

Ibn Shihab (az-Zuhri) said: 'Urwah b. az-Zubayr informed him that 
'A'ishah, the wife of the Prophet (God bless and preserve him), informed 
him that marriage in the Jahiliyah was of four types, (i) One was the 

1 IS, viii. 63-66. 2 Ibid. 109. 3. 3 IH, 759; WW, 282. 

* IH, 548; cf. 351. 5 67. 37. i. 


marriage of people as it is today, where a man betroths his ward or his 
daughter to another man, and the latter assigns a dower (bridewealth) 
to her and then marries her. (2) Another type was where a man said to 
his wife when she was purified from her menses, Send to N. and ask 
to have intercourse with him ; her husband then stays away from her and 
does not touch her at all until it is clear that she is pregnant from that 
(other) man with whom she sought intercourse. When it is clear that 
she is pregnant, her husband has intercourse with her if he wants. He 
acts thus simply from the desire for a noble child. This type of marriage 
was (known as) nikdh al-istibdd\ the marriage of seeking intercourse. 
(3) Another type was where a group (rahf) of less than ten used to visit 
the same woman and all of them to have intercourse with her. If she 
became pregnant and bore a child, when some nights had passed after 
the birth she sent for them, and not a man of them might refuse. When 
they had come together in her presence, she would say to them, 'You 
(pi.) know the result of your acts; I have borne a child and he is your 
(sing.) child, N.' naming whoever she will by his name. Her child is 
attached to him, and the man may not refuse. (4) The fourth type is 
where many men frequent a woman, and she does not keep herself from 
any who comes to her. These women are the baghdyd ( ? = prostitutes). 
They used to set up at their doors banners forming a sign. Whoever 
wanted them went in to them. If one of them conceived and bore a child, 
they gathered together to her and summoned the physiognomists. Then 
they attached her child to the man whom they thought (the father), and 
the child remained attached to him and was called his son, no objection 
to this course being possible. When Muhammad (God bless and pre- 
serve him) came preaching the truth, he destroyed all the types of 
marriage of the Jahiliyah except that which people practise today. 

This description may be taken as accurate so far as it goes, but 
there are certainly gaps in it. It does not tell us, for instance, 
whether in the first type the marriage is virilocal or uxorilocal. 
Nevertheless it is useful to have this classification in mind as we 
consider the various aspects of the question. 

The place of marriage may be dealt with first. In al-Bukhari's 
third and fourth types it is uxorilocal, and probably also in the 
second. This is in accordance with the picture given by pre- 
Islamic poetry. The normal way to begin an ode is to speak of a lost 
love, and one is given the impression of passionate amours, carried 
on with ardour so long as the tribes of the couple are near one 
another, and then ceasing abruptly when the encampments are 
moved. In the most romantic cases the man would only visit the 
woman at night and by stealth; in others he might reside with her 


tribe for a considerable period. This would presumably be a marri- 
age of the third type, but we cannot tell whether the other men in 
it were also of a strange tribe or were of the woman's tribe. The 
second type would also provide an opportunity for a noble 
stranger. The marriage of Hashim in Medina seems to have been 
akin to the desert unions. He spent only a short time with his wife, 
and the son, Muhammad's grandfather, 'Abd al-Muttalib, re- 
mained in Medina and was only with difficulty restored to his 
father's tribe in Mecca. 1 Where a man settled more or less per- 
manently with his wife's tribe, he would often become attached 
to it as a confederate. It is doubtless along this line that we must 
look for the explanation of how al-Akhnas b. Shariq, a confederate 
from at-Ta'if, became head of the clan of Zuhrah at Mecca; as has 
been seen, there are other reasons for thinking that matrilineal 
descent was esteemed in this clan. 2 

It is from Mecca that we have the clearest example of a matri- 
lineal household. In connexion with Muhammad's marriage to 
Maymunah at the 'pilgrimage of fulfilment' in 629/7, we are in- 
formed that Hamzah's daughter 'Ammarah (or Umamah) was 
taken to Medina. Up to this time she had been in Mecca with her 
mother, Salma bint 'Umays. There is no obvious reason for con- 
necting the two happenings, but a little investigation reveals that 
Maymunah bint al-Harith and Salma bint 'Umays were uterine 
sisters, being daughters of a Himyarite woman, Hind bint f Awf ; 
and both Maymunah and Salma seem to have been living in the 
household of Muhammad's uncle al- c Abbas, who was the 
husband of Maymunah's full sister, Umm al-fadl. Thus al- 
' Abbas had a household consisting of his wife and her children, 
together with her two sisters and their children. If the sources are 
right in saying that he gave Maymunah in marriage to Muhammad, 
then he was presumably in charge of the household, though in 
matrilineal systems it is normal for the control to be in the hands 
of the woman's uterine brother or maternal uncle. Moreover not 
all the children of Hind were in this household. A son, Mahmiyah 
b. Jaz', early became a Muslim and went to Abyssinia; there he 
may have been in the household of Asma', the wife of Ja'far b. 
Abi Talib, but he returned to Medina at least two years before 
Ja'far and was given a position of trust by Muhammad. Another 
sister was the mother of Khalid b. al-Walid and presumably 

1 IH, 88; cf. Robertson Smith, 69 f. a Cf. p. 375 above. 


resided with her husband. When Hamzah's daughter 'Ammarah 
came to Medina there was a dispute who should be her guardian; 
Muhammad placed her under Ja'far, less because he was her 
father's brother than because he was husband of her mother's full 
sister. In giving this decision Muhammad quoted a principle ex- 
pressing the unity of a matrilineal group, namely, 'the maternal 
aunt is a mother' (al-khdlah wdlidah). We may conclude, then, that 
this was a group which was largely matrilineal in organization, but 
not entirely so. 1 

Where marriage was uxorilocal and matrilineal ideas predomin- 
ated, a woman of character would have much authority. According 
to Kitdb al-Aghdm, 

the women in the Jahillyah, or some of them, had the right to dismiss 
their husbands, and the form of dismissal was this. If they lived in 
a tent, they turned it round, so that if the door had faced east it now 
faced west, and when the man saw this he knew that he was dismissed 
and did not enter. 

This implies that the tent belonged to the woman or the woman's 
family, and that she allowed him to live in it only so long as she 
pleased. 2 Salma bint 'Amr, the Medinan woman whom Hashim 
married, would only unite with a man on condition that she could 
dissolve the union when she chose. 3 Indeed the women of Medina 
in general were noted for pride and for jealousy of their honour 
and position summarized in the word ghayr. Muhammad is said 
to have remarked that, because of their ghayr he would not marry 
a woman of the Ansar, since she would not have sufficient patience 
to endure fellow- wives ; 4 and, even if this is not the whole reason 
for Muhammad's not marrying a Medinan woman, there is doubt- 
less something in it, and the contrast between the social attitudes 
in Mecca and Medina may explain why there was hardly any inter- 
marriage between the Emigrants and the Ansar. A saying of the 
caliph ' Umar's is recorded : 

we of Quraysh used to dominate (our) women; but when we came 

1 IS, viii. 33, 113, 94-98, 202-9; WW, 302. For Mahmiyah cf. IS, iv/i. 145 f.; 
WW, 177, &c. (disagreeing with IH, 783); also IS, iv/i. 40. 26, iv/2. 8. 20, 37. 17. 
For Khalid cf. IS, viii. 209. 18, and Usd, s.v. 

2 Robertson Smith, 65 ; based on Aghdni, xvi. 106, where Mawiyah dismisses 

3 IH, 88; but cf. her apparently virilocal marriage with Ufcayhah b. al-J 
in Ibn al-Athir, Kdmil, i. 404 f., &c. 

4 IS, viii. 148. 20; Stern, op. cit. 76 f. 


among the AnsSr, they proved to be a people whose women dominated 
them; and our women began to copy the habits of the women of the 
AnsSr. 1 

Despite their great influence, the women of Medina are com- 
monly said to have been unable to own property. This presumably 
means that the property belonged to the matrilineal family and 
was administered by a woman's uterine brothers or maternal uncles 
or sons. This arrangement would not be disadvantageous for the 
woman so long as communal ideas prevailed ; but with the growth 
of individualism the men would sometimes claim that the property 
belonged to them as individuals. As already suggested, the greater 
individualism at Mecca probably helped to make it possible for out- 
standing women like Khadyah and Asma' bint Mukharribah to 
own property and trade in their own name. 

While many matters of detail are to be gleaned from the stories 
of the Jahillyah and biographies of early Muslims, little is said 
about the underlying ideas. It may be helpful, therefore, to quote 
an African parallel. Among certain 'matrilineal' tribes of the 
central Bantu 

there is a remarkable degree of uniformity as to the principles governing 
descent and succession and the various ideologies by which people 
explain their adherence to the mother's rather than to the father's line, 
and stress their community of interests with their maternal relatives. 
Blood is believed to be passed through the woman and not through the 
man. The metaphors of kinship stress the ties between people 'born 
from the same womb' or buckled at the same breast', and in some tribes 
the physical role of the father is believed to be limited to the quickening 
of the foetus already formed in the uterus. 2 

It cannot be assumed that exactly these ideas were found in Arabia, 
especially if revenge for blood was a matter for the patrician. 
Nevertheless there are some close parallels. In the sphere of 
language there is the use of rahim, which properly means 'womb', 
for 'kinship', and of bain or 'belly' for a subdivision of a tribe. 3 
In later Islamic law there are the same forbidden degrees in milk- 
relationship or foster-relationship as in blood-relationship, and 
this justifies an inference to the importance of milk-relationship 
in earlier times among the groups which were chiefly matrilineal, 
since it is essentially a relationship through females. 

1 Al-Bukhari, 67. 83. * Radcliffe Brown and Forde, op. cit. 207. 

3 Cf. Robertson Smith, 28-34, but for ummah see Jeffery, Vocabulary, s.v. 


We cannot tell whether the Arabs believed, like some of the 
Bantu tribes, that the contribution of the father to the heredity of 
the child was slight or negligible, but certainly in some circles little 
attention was paid to the fact of physical paternity. The insistence 
of the Qur'an on the 'iddah or waiting period before remarriage (to 
ensure that the woman has not conceived by her previous husband) 
argues that it was often not observed. That was certainly the case 
in al-Bukhari's third and fourth types of pre- Islamic marriage; and 
even in the second type the precautions to make sure that the child 
was procreated by the stranger sound like an Islamic rewriting of 
an old custom. This would be in accordance with the widely 
accepted principle of 'the child to the bed* (al-walad li y l-firash). 
In other cases also a woman's child was reckoned the son of the man 
who was her husband when the child was born, even when it was 
known that he was not physically the father. 1 Similarly in Islamic 
practice the owner of a slave-woman was the pater of her child, 
even when another man was known to be the genitor. 2 Where 
matrilineal ideas prevailed there was apparently a strong feeling 
that the child belonged to the mother's group. A man called *Ijl b. 
Lujaym, on marrying a pregnant woman, agreed with her previous 
husband to bring up the child and eventually restore it to its 
father's tribe; but when it came to sending away the child, feeling 
in ' IjFs tribe about the wrongness of this procedure was so violent 
that he and the father agreed to abandon the arrangement. 3 

Forms of marriage where little attention is paid to paternity 
sometimes come very near to promiscuity or prostitution, though 
in other cases they may lead to lasting unions. We are so ignorant, 
however, of many of the points involved in Arabian practices, and 
of the interpretation of some of the points which have been re- 
corded, that it is unwise to be dogmatic. Though there is no direct 
evidence, for example, we might suppose that women with infants 
in a matrilineal group made a regular practice of exchanging them 
for the purposes of suckling; the effect of forbidden degrees based 
on milk-relationship would be to make extreme endogamy impos- 
sible. Again, a poet's use of the word kannah to denote his own 
wife, whereas it usually means his sister-in-law or daughter-in-law 
is possibly to be explained as referring to a polyandric group where 

1 Robertson Smith, 109-11. 2 Stern, 93. 

3 Robertson Smith, 115, based on al-Maydam, Majma* al-Amthdl t BftlSq, 
(i867)/i284, i. 160 ( = ed. G. Freytag, Bonn, 1838-43, i. 321). 


brothers cohabited with the same woman; but we cannot be 
certain. 1 

It is difficult, again, to know what was meant by 'adultery* 
in pre-Islamic times. When Hind bint 'Utbah, the wife of Abu 
Sufyan, made her submission to Muhammad along with some 
other women, they were told not to commit adultery, and she 
indignantly replied, 'Does a free woman commit adultery?' 2 To 
a European the natural interpretation of these words would be that 
free women were too proud or too chaste to do such a thing; but 
it is possible that Hind meant that no union a free woman was 
likely to contract was such as to have the term zind applied to it, 
presumably because she had the right to dismiss or at least to 
separate from any husband of whom she tired. In Islamic times 
zind meant 'adultery* in a wide sense, but it is not clear what it 
meant previously. Like other Islamic terms for sexual misconduct, 
it may originally have designated a normal practice in the pre- 
existing social system. The men of at-Ta'if complained to Muham- 
mad that zind was necessary for them since they were merchants. 3 
For them it must have been a practice to which no stigma attached, 
perhaps temporary unions with strangers. 

The loose polyandry which al-Bukhari describes as his fourth 
type of pre-Islamic marriage was another primitive practice which 
a higher civilization was bound to condemn. The word bighd in 
Surat an-Nur (24), 33, is usually translated 'prostitution 1 , and the 
passage is said to refer to certain men Ibn Ubayy is one who is 
named who had control over a number of women and profited 
from their earnings as prostitutes. 4 As they stand, however, the 
stories command little respect. The probability is that bighd refers 
to some form of temporary union with neglect of paternity, in 
which the male head of the household to which the woman be- 
longed shared in the presents given to her possibly al-Bukhari's 
fourth type. This differs from prostitution as understood in the 
West or in modern Islam in that there is no clear line of demarca- 
tion between such unions and others which, according to the 

1 Abu Tammam, Hamdsah, Bulaq, (i87Q)/i296, ii. 33 ( = ed. G. Freytag, 
Bonn, 1828-47, i. 252). 

2 IS, viii. 4. 17, 172. 17; cf. Robertson Smith, 106; Stern, 73. 

3 Stern, ibid. ; cf. Usd, v. 282 story told of Abu Kabir al-Hudhali in explana- 
tion of verses in IH, 646; further references in Farrukh, Das Bild des Fruhislam 
in der arabischen Dichtung, 113; cf. Wellhausen, 'Die Ehe bei den Arabern* 
47* n. 4 Tab., Tafsir, xviii. 93; Wherry (Sale), ad loc.\ &c. 


principles of the matrilineal system of pre-Islamic Arabia, were 
thoroughly respectable. In dealing with all such questions we must 
remember that what is right and proper in one social system may 
be a heinous crime in another. 

(c) Signs of transition 

In pre-Islamic Arabia we not merely find what may be called 
patrilineal and matrilineal groups; there are also indications that 
patrilineal principles were sometimes replacing matrilineal prin- 
ciples. The Islamic religion encouraged patrilineal principles, 1 and 
a century later these were dominant among the Arabs. Neverthe- 
less, the transition had begun before the time of Muhammad. Some 
of the cases already discussed have shown a mixture of the two 
kinds of principle, and others may now be given. 

A story is quoted by Robertson Smith 2 

where to a suitor proposing for a girl's hand the father says, 'Yes, if 
I may give names to all her sons and give all her daughters in marriage.' 
'Nay', says the suitor, 'our sons we will name after our fathers and 
uncles, and our daughters we will give in marriage to chieftains of their 
own rank, but I will settle on your daughter estates in Kindah and 
promise to refuse her no request that she makes on behalf of her people.' 

The suitor here represents patrilineal principles, the father matri- 
lineal but not purely matrilineal, since we should then have 
expected the marriage to be arranged by the maternal uncle of the 
girl. A contrasting story from Islamic times is that of Khansa* bint 
Khidham (or Khudham), apparently of the nomadic tribe of Asad, 
whose first husband was killed at Uhud; her father then tried to 
marry her to another man, but she objected, saying that the 
paternal uncle of her child was preferable, and Muhammad allowed 
her to marry whom she pleased ; the man she chose was a kinsman 
of her first husband, but not actually a brother (though the Arabic 
word 'arnm, rendered by 'paternal uncle' would include this). 3 
Here the woman herself seems to prefer patrilineal principles, and 
once again it is her father who gives her in marriage and who objects 
to patriliny ; but presumably personal matters about which we are 
not informed were also involved. 

1 Cf. ch. VIII, 2, above. 

a Kinship, 102, based on Ibn 'Abd Rabbi-hi, Al-'Iqd al-Farid t BulSq, (i876)/ 

1293, iii- 272. 

3 IS, viii. 334 f., iii/2. 35. i (Asadiyah); Usd, v. 440 f. (Anarlyah), ii. 115 f. 

6783 C C 


Another aspect of the problem of transition is to be scan in the 
complaint made to Muhammad by the women of Medina that they 
were under the domination of their paternal cousins. 1 Somewhat 
similar is the case of the widow of Sa'd b. ar-Rabf ; on his death his 
brother (sc. 'Abd ar-Rahman b. 'Awf who had been bothered' 
with him by Muhammad) assumed control of his property, and 
Sa'd's widow complained to Muhammad that her rights and those 
of her daughters were being infringed and that it would be difficult 
for her daughters to find husbands. 2 Both complaints show that 
patrilineal ideas were gaining ground at Medina. In particular 
cases relatives in the male line (and f Abd ar-Rahman would count 
as such) may have taken over administration of communal property 
which was normally inherited in the female line, because of the 
absence of suitable maternal relatives. The great slaughter at 
Bu'ath and elsewhere would contribute to this absence. For the 
woman, however, control by husband or paternal relatives would 
mean that the property had passed from the control of the matri- 
lineal family with which she primarily identified herself. For the 
proud, 'managing* women of Medina this would mean a loss of 
wealth, prestige, and influence, and, even with the rights conferred 
on them by Islam, they (or the most fortunate of them) would not 
be so secure as they were previously. 

In this connexion the case of Qays b. al-Khatim may be men- 
tioned, though the interpretation of it is obscure. Qays, who was 
a poet of the Medinan clan of Zafar and died before the Hijrah, 3 
married both f lqrab and her daughter (by another man) Hawwa'. 4 
He apparently also looked after his mother, for, when setting out 
to avenge. his father, he left a palm-garden to a kinsman on con- 
dition that, if Qays died, the kinsman would provide for Qays's 
mother (Qarlbah bint Qays of Salimah) 5 from it. 6 There may be no 
connexion between these two facts, but, if so, it is strange that his 
mother was not looked after by his heir (matrilineally) in the usual 
way. If Qays administered the property of her family and lived 

1 Quoted by Robertson Smith, 84, from a manuscript of al-Waljidl on Q. 
4. 19/23. 

2 IS, iii/2. 78, viii. 261 f.; WK, 320-3; WW, 146 f.; cf. Robertson Smith, 
96 f.; Stern, 165; Caetani, i. 569. 

3 Usd, i. 229, s.v. Thabit b. Qays; cf. IH, Index; Ibn Qutaybah, Kitdb ash- 
Shi'r wa-sh-Shiiard\ ed. de Goeje, Index ; &c. 

4 IS, viii. 231, 237. 

5 Ibid. 246. 22. 

6 Aghani, ii. 160; cf. Robertson Smith, 96; Stern, 165. 


in her household, then he must have been afraid that his successor 
as administrator would use the property for his private ends and 
would not treat Qaribah properly. On the other hand, if Qays lived 
with his wives in their household, and looked after his mother 
there, she clearly could not remain in it after his death. In the latter 
case, if Qays was administering the property of his wives, this is 
an instance of the transition to a patrilineal system. There may 
have been patrilineal tendencies in the wives' family, since 'Iqrab's 
brother Sa'd seems to have inherited the power of his father, 
Mu'adh b. an-Nu'man. 

One wonders to what extent cases where a man married two 
women of the same matrilineal household are signs of a movement 
towards patriliny. There are some examples of marriage with two 
sisters from both Mecca and Medina, 1 though we are not informed 
whether they were contemporary or successive. Such a position 
would presumably give a strong man much influence, especially 
if the women's relatives on the female side were weak. It is also 
not clear whether quarrels between a woman's husband and her 
brother are specially connected with the transition from matriliny 
to patriliny, or whether such quarrels are always to be found in 
a matrilineal society. The instances of Jassas killing his sister's 
husband Kulayb, 2 and of Salma bint c Amr (mother of 'Abd al- 
Muttalib) helping her brothers against her husband, 3 are con- 
siderably before the Hijrah. 

(d) Conclusions 

A little investigation of pre-Islamic ideas about marriage and the 
family shows how much remains obscure and requires further 
study. Unfortunately, in order to assess Muhammad's achievement 
in this sphere, it is necessary to adopt some view of the nature of 
what he found. The view is therefore suggested here by way of 

1 Abu Uhayhah Sa'id b. al-'A? ('Abd Shams) Hind bint al-Mughirah, 
afiyah (MakhzOm); ash-Shahrastani, Kitdb al-Milal tva 'n-Nihal, ed. 
Cureton, p. 440. 

Qays b. Makhramah (al-Muttalib) Waddah, Umm Sa'd ('Abd al-Ash'hal); 

IS, viii. 232. 
'Amr b. tfarSm (Salimah) ar-Ribab bint Qays, Hind (Salimah); IS, iii/2. 

105, 119. 2, viii. 288. n. 
Mu'adh b. 'Amir (Pubay'ah) ; Layla bint Abi Sufyan, 'A'ishah (Pubay'ah); 

IS, viii. 253(?)- 

2 Cf. Caussin de Perceval, ii. 276-8; Nicholson, Literary History, 56. 
5 Ibn al-Athir, Kdmil, i. 404 f. 


hypothesis that a transition was in progress from a matrilineal 
system to one that was wholly or largely patrilineal, and that this 
transition was linked with the growth of individualism. This seems 
more likely in itself than the alternative view that from time 
immemorial some tribes had been matrilineal and some tribes 
patrilineal, and it also fits in with the standpoint of this book which 
sees in the growth of individualism from perhaps about the middle 
of the sixth century an important cause of the malaise with which 
the religion of Islam dealt. No strict proof can be given of the view 
adopted, but it is not improbable as an explanation of the facts, 
and it leads to a reasonable account of the reforms achieved by 

The important question is not where the marriage takes place, 
for uxorilocality can exist in a patrilineal system, and may have 
continued for some time after Muhammad's reforms. The essential 
point is the composition of the group which owns and inherits 
property. Until the later sixth century, we assume, the group was 
a matrilineal one, and the property was held communally, or at 
least as a trust for the common good. It would normally be adminis- 
tered by the uterine brothers of the women concerned. When 
individualism appeared, however, and men began to think of them- 
selves more as individuals than as members of a group, and to set 
private interests above the interests of the group, there would be 
a tendency for a man to appropriate for personal use as much as 
he could of the communal property he administered. This would 
still be within the matrilineal system, of course. A transition to the 
patrilineal system would come, however, if a man tried to hand 
on this property he had appropriated not to his sister's son but to 
his own son. There are no doubt also other ways in which the 
transition could come about. When there were many deaths (as 
had been the case in Medina just before the Hijrah), one man might 
have to look after the property of several families ; and as many 
marriages were between closely related families, people would often 
stand in several different relations to one another, and the strong 
man would know how to benefit from this confusion. 

While we can only conjecture about the precise manner in which 
the change from a matrilineal to a patrilineal system was effected, 
it is certain that it was in progress before the Hijrah, and that 
in the century after the Hijrah the matrilineal system largely 


The Technical Terms in Surahs 4. 24/28, 5. 5/7, and 24. 33 

THESE three passages are important for the understanding of 
Muhammad's attitude to some of the uxorilocal marriage customs 
of Medina, but the original meaning of the terms in them has been 
lost owing to centuries of reinterpretation to make them fit later 
Muslim practice. Even the dictionaries are of little help, since the 
meanings they give are those attributed to the words after this 
process of reinterpretation. The passages, with the crucial words 
merely transliterated, are as follows: 1 

(4. 24/28) (Forbidden to you are) . . . and the muhsindt among the 
women [except what your right hands obtained the rescript of God 
for ypu]. Allowable for you with regard to what is beyond that is the 
seeking them by your wealth, but muhsinln and not musdfihtn. And for 
what you enjoy from these women (v. I. adds 'up to a fixed term') pay 
them their hire as stipulated (or 'at the stipulated time'). But with 
regard to what you mutually agree on after the stipulated time, no sin 
is attributed to you ; God is knowing, wise. 

(4. 25/29) And he of you who has not a superabundance so as to 
marry believing muhsindt, (may take) [of what your right hands possess] 
of your believing fataydt God knows your faith, that you are one of 
another. Marry them with the consent of their people, and give them 
their hire reputably, they being muhsindt, not musdfihdt and not taking 
akhddn. (25 cont./3o) And when, after ihsdn ( ? = becoming muhsindt), 
they commit fdhishah, their penalty is half that of the muhsindt. . . . 

(5. 5/7) Today there are made allowable for you the good things; . . . 
also the muhsindt of the believers and the muhsindt of those who were 
given the Book before you, if you give them their hire, muhsintn, not 
musdfihin and not taking akhddn. 

(24. 33) Let those who do not find (means) to marry be continent 
until God enriches them of His bounty. For those your right hand 
possesses who desire the writing (of manumission) write it if you know 
any good in them, and give them of God's wealth He has given you. 
Do not compel your fataydt to bighd* if they want tahassun, in order that 
you may seek the gain of this present life. . . . 

The first word to attract attention is muhsindt. The usual text is 

1 Round brackets indicate additions made in translation to elucidate the 
meaning; square brackets indicate portions of the actual text which I take to 
be later additions to the original text. 


muhfandty this being the passive participle, whereas the other is 
the active participle. There would seem, however, to be a close 
parallel between 'muhstnat, not musdfihdf and *muh$inin, not musd- 
fihm 9 ; but in the latter phrase it is always the active of the masculine 
which is found, and it is therefore most likely that this would be 
the form of the feminine also. Lane gives such meanings as the 
following: continent, chaste, abstaining from what is not lawful 
nor decorous, married, having a husband, emancipated, having 
become a Muslim woman; the passive form would indicate that it is 
her husband who has caused her to be continent, chaste, &c. The 
basic meaning of the root is to be inaccessible or unapproachable, 
as a fortress, and this perhaps links up with the idea of chastity. 
The meaning of 'married* is probably derived from the beginning 
of the first Qur'anic quotation. 

Now the act or state expressed by muhsinln or muhsindt is 
approved by the Qur'an, for both men and women apparently, 
while the opposite is disapproved. In the light of what we know 
about the existing situation and the Islamic reforms the most 
satisfactory meaning would be 'observing purity of paternity*. This 
would mean that a woman observed an *iddah or waiting-period 
after separating from a 'husband* before having intercourse with 
another man; where the woman was pregnant the 'iddah would 
last until after the birth of the child at least, and perhaps until she 
had stopped suckling it. In short, this is the restriction of a woman 
to one man at a time, and may be referred to as monandry, while 
in some contexts muhsindt could be rendered as 'monandric*. In 
the first instance of the word above it must be taken as involving 
not merely the general practice of monandry, but a marriage 
actually in force. There is no difficulty about applying this concep- 
tion of monandry to the other instances both of the participles and 
of the noun tahassun. A precise meaning of this kind is more likely 
to be the original meaning than one of the vaguer meanings com- 
monly given to the words. Other precise meanings are conceivable 
perhaps linking up with virilocal marriage but the one given 
best fits all the instances. 1 

The opposing types of action, represented by musdfihln and 
musdfihdt, may be described as 'polyandric', though of course there 

1 For the modern legal meaning cf. Snouck Hurgronje in Zeitschrift der 
deutschen morgenldndischen Gesellschqft, 53. 161 f. = Verspreide Geschriften, ii. 

TECHNICAL TERMS IN SORAHS 4.24/28, 5. 5/7, 24. 33 391 

are other sorts of polyandry. In Medina this would mean that a 
woman had intercourse with a number of men without observing 
any 'idddh, so that there could be no certainty about the father of 
any child she had. In the dictionaries sdfaha is given the vague 
meaning of committing fornication. The basic meaning of the first 
stem of the word is to pour out (a liquid) or to spread (as a camel 
spreads itself on the ground). Thus sdfaha is not inappropriate as 
a description of the 'unlimited* polyandry mentioned by al-Buk- 
hari. The complementary phrase, 'not taking akhddn\ confirms 
this, for akhddn are 'secret or private friends', and after 'visits' from 
such persons a woman would be unlikely to observe an 'iddah. In 
its outward aspect the practice is hardly distinguishable from 
prostitution (at least in extreme cases), but it differs in its social 
setting and in the underlying ideas. The word bighff in 24. 33 
presumably refers to the same practice. 

The fataydt in 4. 25/29 and 24. 33 are commonly said to be 
slaves, and the word certainly can refer to female slaves, though 
its primary meaning is young women. But the command to 'marry 
them with the consent of their people' (ahl) would imply that they 
belong to someone else, and regular marriage (nikdh) with another 
man's slave at this period is most unlikely. The words 'of what 
your right hands possess', which support the traditional interpreta- 
tion, seem to be an addition made some time after the original 
revelation of the passage, presumably when female captives were 
plentiful; but the original remark about 'consent of their people' 
was not expunged. It would be more satisfactory to regard the 
fataydt as a class of women in matrilineal households, possibly 
younger sisters and more distant kinswomen of the leading woman, 
who were not allowed to have a permanent husband resident in 
the household, but were allowed to receive 'visits' from men as 
they pleased. That they were a distinct class seems clear from the 
lighter punishment prescribed in 4. 25/30. Since a marriage with 
one of these women was to cost less, the man presumably went to 
live ( ? for a time only) in the woman's household, and while she 
was married to him she was not to have intercourse with any other 
men; he may indeed have paid only a series of 'visits' to her. The 
lighter penalty may be due either to the woman's recent change 
of status from 'polyandric' to 'monandric', or to her continued 
residence in her old home with the temptation to revert to the old 
ways. The word fdhishah, meaning something like 'an abominable 


thing', appears to be applied not to adultery specifically but to any 
practice which was formerly normal but is now regarded as objec- 
tionable by Islam; in 4. 22/26 marriage with one's stepmother is 

If this treatment of the technical terms is sound, the passages 
under discussion may be interpreted somewhat as follows. The 
first (4. 24/28) gives permission to contract a temporary alliance 
with a 'polyandric' woman, provided she gives up her polyandry. 
The following verse is similar, but the alliance is not necessarily 
temporary and the woman's people must consent so that it cannot 
be secret; this seems to be a little later than the previous verse. 
The next passage is perhaps earliest of the four, since it speaks of 
'people of the Book', presumably Jews, and permits marriage with 
'monandric' women from these and from the believers generally. 
The last passage illustrates how vested interests try to perpetuate 
the existing system. 

Muhammad 9 s Marriages 

THE classification of Muhammad's marriages has been influenced 
by a verse of the Qur'an, 33. 50/49. Unfortunately there is some 
difficulty in the interpretation and even the translation of this 
verse. In the century after Muhammad's death conditions changed 
so much that men had ceased to be aware of many of the facts of 
the social system to which this verse was relevant. It must therefore 
be examined in detail. A rough translation is as follows (the clauses 
being numbered for convenience) : 

O prophet, We have made allowable for thee (i) thy wives whose 
hires (ujur) thou hast given, (2) and what thy right hand has possessed 
of the. booty God has bestowed on thee, (3) and the daughters of thy 
paternal and maternal uncles and aunts, (4) those who emigrated with 
thee, (5) and a believing woman, if she gives herself to the prophet, if 
the prophet wants to marry her, (6) being special for thee apart from 
the believers. 

The first group (i) consists of Muhammad's wives in the strict 
sense. The word 'hires' (ujur) is usually interpreted as Mowers' 
(sing. mahr). It must be either mahr or something which took the 
place of mahr in the early days. It might perhaps be the annual 
supply of provisions from Khaybar. 

The second group (2) are the slave-concubines, women who had 
been captured in war and not set free (like Juwayriyah). Rayhanah 
is sometimes said to have belonged to this group. So presumably 
did Mariyah the Copt, though she was not captured but pre- 

The third group (3) is of cousins. It is not clear whether the 
following clause (4), 'those who emigrated with thee', is a limitation 
on the right to marry cousins or signifies a fourth group. 'Abdallah 
b. Mas'ud read 'and' before the phrase and thus made it clear that 
this was a distinct class ; to such a class one might assign Zaynab bint 
Khuzaymah, who presumably emigrated to Medina with her hus- 
band, e Ubaydah b. al-Harith of the clan of al-Muttalib. A story is 
told about Muhammad's cousin, Umm Hani', a daughter of Abu 
Talib, which presupposes that clauses (3) and (4) indicate only one 
class; she refused a proposal of marriage on the grounds that she 


had not emigrated with Muhammad. 1 This is not infallible evi- 
dence, however, since there are other accounts of her refusal to 
marry Muhammad which do not mention this point. The solution 
of the problem of clauses (3) and (4) depends on the view adopted 
of the purpose of the rule stated in them. The most likely view is 
that these women, whether two groups or one, did not receive ujUr 
like the first class; and the reason for this would presumably be 
either that Muhammad had already provided for them (as his 
cousins or as widows of his community), 2 or that he was their 
official male representative and did not hand over property to 
himself. The marriage of cousins was not prohibited in Islam, and 
the verse could not be a prohibition to Muhammad to marry 
further wives, 3 since at the time of his death he was in the process 
of arranging a marriage with Qutaylah bint Qays. 4 

The last group (5) is that of believing women who gave them- 
selves to Muhammad. These are doubtless women who contracted 
a union with Muhammad according to the old principles (perhaps 
as modified in Q. 4. 25/29). 5 Various women are mentioned as 
belonging to this category. With the exception of Maymunah none 
of them seems to have had an apartment in Muhammad's residence 
in Medina. It is interesting that Maymunah is included in this 
group, since she came of a matrilineal family. 6 In some accounts 
al-'Abbas is said to have arranged her marriage, but the alternative 
report, that she Entrusted her affair' to Muhammad may mean 
that al-'Abbas had only a secondary part in the arrangements. 7 
One wonders whether when she first went to Medina along with 
her niece, Hamzah's daughter, the two of them lived in the house- 
hold of her sister and Ja'far b. Abi Talib. When her marriage to 
Muhammad was consummated during his return journey to 
Medina from the * pilgrimage of fulfilment' ('umrat al-qadiyah) 
she had her own tent (qubbah) ; 8 but this was perhaps usual in the 
case of Muhammad's wives and may not signify that the marriage 
was uxorilocal. Apart from Maymunah, however, Muhammad's 
unions with 'believing women who gave themselves to him' were 

1 IS, viii. 109. 9-12; cf. 15-19. 2 As for Umm Hani', ibid. 32. 13. 

3 Contrast Tab., Tafslr t xxii. 15 top. 

4 IS, viii. 105. 5 Cf. p. 391 above. 

6 Cf. p. 380 above; for her giving herself cf. IS, viii. 98. 3. 

7 IS, viii. 95. 3; for the phrase ja'alat amra-hd ild cf. ibid. 95. 6 (Maymunah 
to al-'Abbas), 82. 15 (Zaynab bint Khuzaymah to Muhammad). 

8 IS, viii. 100. i. 


presumably uxorilocal. This would be in line with the current 
practice in those sections of Arabian society where matrilineal 
principles were dominant. 

There remains the interpretation of the last clause (6), khdlisatan 
la-ka min dun al-mu'mimn. These words are now always taken to 
mean '(this is a) special (privilege) for you as distinct from the 
believers' ; the special privilege is often held to be marriage with 
a woman who gives herself to Muhammad without a wait and 
without a 'dower'. 1 This interpretation is in harmony with the 
attempt by 'Umar and others to stop mufah unions, and could be 
a device to bolster up the prohibition of mufah. While this inter- 
pretation is not impossible, a more natural one would be to take 
khdlisatan as an adjective qualifying 'woman', so that the clause 
would mean '(keeping herself) special for you and not (having 
sexual relations with other) believers'. This would be in line with 
the insistence on women becoming 'monandric'. 

The categories of women named in this verse have had some 
influence on the later accounts of the wives of the prophet. The 
phrases 'bestowed by God as booty' and 'gave herself to the 
prophet' frequently occur. In his eighth volume Ibn Sa'd, in 
addition to the 'wives of the prophet', has lists of 'women whom 
the Messenger of God married without consummating the marri- 
age, and those whom he divorced* and 'women to whom the 
Prophet made a proposal without completing the marriage, and 
women who gave themselves to the Messenger of God'. Some of 
the subsumed cases do not fit the headings. Thus there is no word 
of Muhammad wanting to marry Umamah bint Hamzah; someone 
else suggested the match, and Muhammad said that it was im- 
possible since Hamzah was his brother by fosterage. 2 

Muhammad is usually said to have had fourteen wives in the 
strict sense, of whom nine survived him; but there is some dispute 
about the identity of the fourteen. 3 The following is a list of the 
women whom he married or with whom he contemplated marriage. 

1. Khadijah bint Khuwaylid (Quraysh Asad), married about 
595 when she was aged 4O. 4 

2. Sawdah bint Zam'ah (Quraysh 'Amir), married about 620, 
aged about 30 (?); widow of as-Sakran b. 'Amr, an early Muslim, 

1 Cf. Qatadah ap. Tab., Tafstr, xxii. 14. 24 f. a IS, viii. 113 f. 

3 Ibid. 156-9; IH, 1001-5 names those in the first list together with Asmft', 
'Amrah, and Umm Sharik. 

4 IS, viii. 7-1 1, i/i. 85; Tab. i. 1766 f.; Caetani, i. 166-73. 


with whom she made the hijrah to Abyssinia and returned to Mecca ; 
his brother was a prominent pagan, and her own brother remained 
in Abyssinia. 1 

3. 'A'ishah bint Abi Bakr (Quraysh Taym), married in 623/1 
aged 9; 2 the only virgin Muhammad married. 

4. Hafsah bint 'Umar b. al-Khattab (Quraysh 'Adi), married 
in 625/3 a g e d *8; widow of a Muslim killed at Badr. 3 

5. Umm Salamah (Hind) bint al-Mughirah (Quraysh Makh- 
zum), married 626/4 a g e d 2 95 ^ er husband Abu Salamah had died 
of wounds received at Uhud. 4 

6. Zaynab bint Khuzaymah ('Amir b. Sa'sa'ah), married 626/4 
or the previous year, aged about 30, and died a few months later ; 
after a divorce from at-Tufayl b. al-Harith (Quraysh al-Muttalib) 
she had married his brother 'Ubaydah who was killed at 

7. Juwayriyah (al-Mustaliq of Khuza'ah), daughter of the chief 
of the tribe, captured in the attack on it in January 627 (viii/5), 
married by Muhammad on her profession of Islam and set free ; 
aged 20 at marriage; perhaps only a concubine at first, but before 
his death had become a full wife. 6 

8. Zaynab bint Jahsh (Asad b. Khuzaymah), married Muham- 
mad in 627/5 a f ter her divorce from Zayd b. Harithah, aged 38; 
her mother was a maternal aunt of Muhammad's, and her father 
a client of the clan of 'Abd Shams of Quraysh.* 

9. Mariyah the Copt, a slave-concubine presented to Muham- 
mad by the ruler of Egypt in 628/6 or earlier, who bore him a son 
called Ibrahim ; she remained a concubine. 8 

10. Umm Habibah (Ramlah) bint Abi Sufyan (Quraysh 'Abd 
Shams), married on Muhammad's return from Khaybar in 628/7, 
aged about 35; she was the widow of 'Ubaydallah b. Jahsh, with 
whom she had made the hijrah to Abyssinia. 9 

1 1 . Saf lyah bint Huyayy (Jewish an-Nadir), captured at Khay- 
bar in 628/7 an d assigned to Muhammad; aged 17; was perhaps 

1767-9; Caetani, i. 312. 
1769 f.; Caetani, i. 424. 
1771; Caetani, i. 540. 
1771; Caetani, i. 588 f. 

IS, viii. 35-39; Tab. 
IS, viii. 39-56; Tab. 
IS, viii. 56-60; Tab. 
IS, viii. 60-67; Tab. 

IS, viii. 83-85; Tab. 

8 IS, viii. 153-6; Tab. 

IS, viii. 82; Tab. i. 1775 f.; Caetani, i. 588 f. 

1772; Caetani, i. 60 1. 

IS, viii. 71-82; Tab. . 1772 f.; Caetani, i. 610 f. 

1775; Caetani,i. 730. 

IS, viii. 68-71; Tab. . 1772; Caetani, ii. 55. 


a concubine at first, but apparently accepted Islam and was set 
free. 1 

12. Maymunah bint al-Harith ('Amir b. Sa'sa'ah Hilal), 
married as Muhammad returned from the 'pilgrimage of fulfilment* 
in 629/7, aged 27; sister of the wife of al-'Abbas, &c. 2 

13. Rayhanah bint Zayd (Jewish an-Nadir), captured with 
Banu Qurayzah to which her husband belonged, in 627/5 > became 
Muhammad's concubine, and apparently retained that status ; died 
before him in &32/IO. 3 

These are the women who may be regarded as having been 
properly united to Muhammad as wives or concubines. At his 
death three were already dead, and Mariyah was only a concubine. 
The remaining nine became the 'mothers of the believers'. About 
a score of other women are mentioned as having been at least 
thought of as wives for Muhammad. There is much obscurity and 
dubiety about some of them ; many tribes were doubtless eager to 
claim a matrimonial relationship with Muhammad, and to make 
the most of vague reminiscences. Thus it is widely held that 
Muhammad married a woman of Kilab, but several completely 
different versions are given of her name. The one thing that seems 
certain about this supplementary list is that none of the women 
in it formed a lasting union with Muhammad. 4 

1. Asma' bint an-Nu'man (Kindah Jawn), said to have been 
married to Muhammad in June 630 (iii/9) and according to some 
versions divorced before the marriage was consummated (but the 
story of the divorce is told of several other women) ; also said to 
have observed the veil and been counted as one of the 'mothers of 
the believers', but this is denied by some accounts, and she is said 
to have married a husband after leaving Muhammad. 5 

2. Qutaylah bint Qays (Kindah), sister of al-Ash'ath b. Qays 
who revolted against Abu Bakr but later became an important 
Muslim leader; was on her way to marry Muhammad when he 
died. 6 

3. Mulaykah bint Ka'b (Layth), said to have been divorced from 

1 IS, viii. 85-92; Tab. i. 1773 ; Caetani, ii. 34 ff., 49. 

2 IS, viii. 94-100; Tab. i. 1773; Caetani, ii. 66. 

3 IS, viii. 92-94; Tab. i. 1775; Caetani, i. 634, ii. 369. 

4 Cf. G. H. Stern, Marriage in Early Islam, 151-7; Caetani, ii. 478 f.; Well- 
hausen, 'Die Ehe bei den Arabern', 464 f. 

5 IS, viii. 102-5, 158. 13, 25; Tab. i. 1775. 3; IH, 1004 f. 

6 IS, viii. 105 f., 158. 16; Tab. i. 1776. 5; IH, 1004. 15 (?). 


Muhammad before consummation of the marriage, or to have 
married him in January 630 (ix/8) and then died; it is also denied 
that Muhammad married any woman of Kinanah (of which Layth 
was a part). 1 

4. Bint Jundub b. Damrah (Kinanah) ; her marriage to Muham- 
mad both asserted and denied. 2 

5. Fatimah bint ad-Dahhak (Kilab), one of the names given for 
the 'woman of Kilab' who is generally agreed to have been among 
Muhammad's wives and to have been divorced; the date of the 
marriage is given as March 630 (xi/8). 3 

6. 'Amrah bint Yazid (Kilab), perhaps variant of above. 4 

7. 'Aliyah bint Zabyan (Kilab), another variant. 5 

8. Saba bint Sufyan (Kilab), another variant. 6 

9. Nashah bint Rifa'ah (Kilab), another variant; but it is also 
said that her clan were confederates of the Jewish clan of Qurayzah. 7 

10. Ghazlyah bint Jabir, Umm Shank (Kilab or Kindah or 
Quraysh-'Amir or Daws or Ansar) ; there is a wide agreement that 
the fifth clause of the Qur'anic verse discussed above (about the 
believing woman who gives herself to the prophet) refers to a 
woman called Umm Shank, and she is sometimes identified with 
Ghazlyah; but there is another version according to which the 
proposal came from Muhammad, and Ghazlyah was divorced 
before the marriage was consummated. 8 

11. Fatimah bint Shurayh; perhaps a corruption of another 
name since no details are given. 9 

12. Sana or Saba bint (Asma' b.) as-Salt (Sulaym), died before 
her marriage to Muhammad was consummated. 10 

13. ash-Shanba' bint 'Amr (Ghifar, confederates of Qurayzah, 
or Qurayzah), divorced because she made a sceptical remark on the 
death of Muhammad's son Ibrahim. 11 

14. Khawlah bint al-Hudhayl (Taghlib), niece of Dihyah b. 
KhaJifah al-Kalbl, married to Muhammad, but died on her way 
to him. 12 

1 IS, viii. 106, 158. 8, ii. 2 IS, viii. 106. 

3 IS, viii. 100. 24, 101. 6, 158. 9, 26. 

4 IS, viii. 100. 25, 102. 7; Tab. i. 1777.3; IH, 1004 f. 

5 IS, viii. 100. 26, 1 02. 10; Tab. i. 1776. 2. 

6 IS, viii. loi. i, 102. 5. 7 Tab. i. 1774. 3. 

8 IS, viii. 110-12; Tab. i. 1774. 15, 1776. 7; IH, 1004. 3. 

9 Tab. i. 1776. 7. 10 IS, viii. 106 f.; Tab. i. 1774. 6. 
11 Tab. i. 1774. 10. " IS, viii. ii4f.; Tab. i. 1776. 12. 


15. Sharaf bint Khalifah (Kalb), maternal aunt of Khawlah, 
took her place on her death. 1 

1 6. Khawlah bint Hakim (Sulaym), daughter of a woman of the 
Meccan clan of 'Abd Shams and related to the clan of Hashim, 
'was one of those who gave themselves to the prophet, and he put 
her off (arjcta-ha) and she used to serve him', this being presum- 
ably after the death of her husband 'Uthman b. Maz'un about the 
time of the battle of Uhud (she had been long married to 'Uthman 
since her son as-Sa'ib b. 'Uthman fought at Badr). 2 (Perhaps, as 
she was the widow of an early Muslim, Muhammad gave her shelter 
under his roof, but refused to marry her because she was too old 
or for some other reason.) 

Finally there are seven women between whom and Muhammad 
there was some talk of marriage without the plans ever being carried 
out. Two were women of the Ansar who arranged the marriages 
themselves but were forced by their families to abandon them, 
perhaps through fear that Muhammad would cease to be impartial. 

1. Habibah bint Sahl (Ansar Malik b. an-Najjar). 3 

2. Layla bint al-Khatim (Ansar Zafar). 4 

3. Umm Hani* bint Abi Talib (Quraysh Hashim). 5 

4. Umm Habib bint al-'Abbas (Quraysh Hashim). 6 

5. Duba'ah bint 'Amir ('Amir b. Sa'sa'ah). 7 

6. Safiyah bint Bashshamah (Tamim al-'Anbar). 8 

7. 'Ammarah (or Umamah) bint Hamzah (Quraysh Hashim). 9 

1 IS, viii. 115; Tab. i. 1776. i. 2 IS. viii. 113. 

3 IS, viii. 326 f. 4 IS, viii. 107 f.; Tab. i. 1776. 14. 

5 IS, viii. 108 f.; Tab. i. 1777. 5. 

6 Tab. i. 1777. 15. 7 IS, viii. 109 f.; Tab. i. 1777. 7 
8 IS, viii. no; Tab. i. 1777. 13. 9 IS, viii. 113 f. 


Names of authors occurring in the footnotes are included in the index only where 
the footnote contains bibliographical information. Names on the map on p. 152 
are not indexed. 'B.' after a name is short for Banu, 'sons of', and indicates a 
tribe or clan. Where the name of a man's clan or tribe appears after his name, it is 
not usually indexed separately, but accompanies the man's name in the index. 
The article a/-, ad- t &c., is disregarded in the alphabetical ordering. 

Abn b. Sa'id ('Abd Shams), 59, 341. 
Abbad (or 'Abd), 131. 

b. Bishr ('Abd al-Ash'hal), 367. 

b. Nahik (Harithah), 376. 

al- 'Abbas b. 'Abd al-Muftalib, 60, 64, 
69 f., 249, 256, 288, 380, 394, 397- 

b. Mirdas (Sulaym), 74, 96 f., 
349 f-, 352 f. 

'Abbasids, 60, 225. 

Abbott, Nabia, 186. 

'Abd (Quraysh), B., 69. 

'Abd b. 'Adi (Kinanah), B., 356. 

*Abd 'Amr, 135, 310. 

'Abd 'Amr b. ayfi, 179. 

'Abd al-Ash'hal b. Jusham, B., 153 f., 

156-64, 168-72, 174-6, i8of., i94f., 

210 f., 214, 216, 227. 
'Abd al-*Aziz, 329. 
'Abd ad-Dar, B., 22, 56-58, 74. 
*Abd yarithah, B., 154. 
'Abd al-Ka'bah, 135, 310. 
'Abd al-Malik, u. 
'Abd Manaf of Quraysh, B., 1 1 1. 
'Abd Manat, B., 20, 138. 
'Abd al-Masih, 127. 
'Abd al-Muftalib, 9O , 263, 375, 380, 


'Abd al-Qadir, 206. 
'Abd al-Qays, B., 82, 131 f. 
'Abd ar-Rafoman, 310. 
'Abd ar-Rahman b. 'Awf, 44, 115, 135, 

249-53, 260, 288, 310, 331, 341, 

344, 37i, 386. 
'Abd Shams (Quraysh), B., 5, u, 50, 

56 f., 73, 87 f., 95 f-, 238, 287 f., 

396, 399- 

'Abd al-'Uzza, 310. 
'Abd Yalil b. 'Amr, 103. 
'Abdallah b. 'Abdallah b. Ubayy, 186. 

b. 'Abd al-Muftalib, 375. 

b. Abi Rabi'ah, 56, 62, 67. 

b. Abi Umayyah (Makhzum), 60. 

b. 'Amr b. rjaram, 180. 

b. 'Amr (Salimah), 250. 
- b. 'Atiq, 213. 

b. 'Awsajah al-'Uram, 99. 

Dhu '1-Bijadayn, 310. 

(b. Hilal) b. al-Khafal, 68. 

5788 D d 

'Abdallah b. IJudhafah as-Sahmi, 345. 

b. Jahsh, 5-7, 9, 340. 

b. Jubayr, 22, 25. 

b. Jud'an, no, 344. 

b. Ka'b al-Mazini, 236. 

b. Mas'ud, 350, 393. 

b. ar-Rabi c , 89. 

b. Rawahah (Ba'1-Harith), 54, 167, 
180, 213, 236, 321, 341. 

b. Sa'd b. Abi 8arfc, 68. 

b. $afwan, 138. 

b. Sallam (Qaynuqa*), 197, 206, 

214, 317. 

b. Ubayy, 15, 21 f., 156 f., 166, 
i68f., 174 f., 194 f., 197, 228, 378; 
see also Ibn Ubayy. 

b. Unays, 30 f., 85, 90, 213, 340. 

b. Zayd (Ba'1-Harith), 3 57, 366. 

b. az-Zubayr, 352. 
al-'Abir, 310. 

al-Abjar, B., 154, 160, 167. 
al-Abna', 118, 129 f. 

(Tamim), 139. 
Abrahah, 90. 

Abraham, 204-7, 303 f., 311, 314, 358. 
'Abs, B., 91, 94. 
Abu, 374. 

'Afak ('Amr b. 'Awf), 15, 18, 
178 f., 340. 

'Amir ar-Rahib, 36, 163, 175, 179, 
190, 378. 

'Amr, 377. 

Abu 'l-'As b. ar-Rabi', 45. 

'1-Bakhtari, 14. 

Abu Bakr, 63, 80, 86 f., 93, 98-101, 
121 f., 130, 186, 197, 250, 254, 
287 f., 294, 307, 33i, 342-4, 36o, 
366, 397. 

Abu '1-Bara' 'Amir b. Malik, 3 1 f., 97 f. 

Abu Basir 'Utbah b. Usayd, 61-63, 

Dharr, 67, 84. 

Dubays, 87. 

Dujanah, 212. 

Abu '1-Harith b. 'Alqamah (Rabi'ah), 

'1-Haytham b. at-Tayyihan, 159, 
1 80. 



Abu Hudhayfah, 344. 

Ja'far an-Nahfras, 9. 

-Jahl, 10-14, 56 f., 137, 263, 311, 

Jandal, 59. 

Jarwal, 99. 

Kabir al-Hudhali, 384. 

Kabshah, 344, 374. 

Lubabah b. 'Abd al-Mundhir, 163, 
188-90, 214, 234, 374- 

Ma'qil b. Nahik (Harithah), 376. 

Mu'ayt, 377. 

Musa '1-Ash'ari, 122, 366. 

Na'ilah, 18, 210. 

Qatadah ar-Rib'i, 342. 

Qays b. al-Aslat (Wa'il), is6f., 
165, 175, 178 f., 377- 

Rabi'ah, 376. 

Raft' Sallam b. Abi'l-Huqayq, 31, 
34,211-13, 340. 

Ruhm al-Ghifarl, 84, 246. 

Sa'd b. Wahb, 197. 

Sa'id al-Khudri, 167. 

Salamah, 30, 340, 378, 396. 

Salih, 304. 

'sh-Shahm, 216 f. 

Sufyan b. Harb('Abd Shams), iof., 
14, 17, 19 f., 23, 25, 28 f., 34, 56-60, 
62-66, 69f., 73-75, 85, 103, 108, 
123, 288, 331, 349, 351 f., 360, 367, 


Sufyan b. al-Harith b. 'Abd al- 
Muftalib (Hashim), 60. 

urad, 99. 

TSlib, 67, 375, 393- 

Tammam, 232. 

.TharwSn, 99. 

'Ubaydah, 52, 341 f. 

Uliayhah Sa'id b. al-'A ('Abd 
Shams), 275, 374, 387. 

'Uzayr ('Abd ad-Dar), 57. 

Yiisuf, 128. 
al-Abwa, 84, 340. 
Abyssinia(-ns), 41, 118, 127, 143, 239, 

247, 288, 293, 305, 315 f- 319, 323, 
343~6, 35i, 375, 380, 396. 

Acts, 200. 

'Atfal, B., 81, 88, 90. 

Adam, 335. 

'Adawl, 108. 

'Adhra'at, 209. 

Adhruh, 115-17. 

'Adi (Quraysh), B., n, 57, 163. 

b. Abi Zughba', 85. 

b. Hatfm (Jayyi'), 89, 367. 

b. an-Najjar (Khazraj), B., 154, 
159, 1 66. 

'Afrfi', 166, 378.